Eurorack (?) Module Power Filter / Conditioning Capacitor(s)

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: luketeaford, Joe., lisa, Kent

Post Reply
User avatar
maniac303
Learning to Wiggle
Posts: 14
Joined: Sun Oct 14, 2018 9:35 pm
Location: Vancouver

Eurorack (?) Module Power Filter / Conditioning Capacitor(s)

Post by maniac303 » Mon Nov 19, 2018 1:51 am

Question: Some module builders/designers condition the provided 12v power further, right on the module, with zener diode followed by ferrite bead, followed by a string of parallel caps. This pattern is on both plus and minus 12v rails with the caps tied to ground.
Question is....why use several small (100nf) caps in parallel, (followed by a larger (10 uF) polarized cap)? Why not just use a single bigger cap, followed by the even bigger polarized cap? Is there an advantage to using a handful of different devices? Is there something else going on here that I don’t understand? For example see Befaco modules where they use 10 or 11 caps where 2 would do?
Second question for bonus points, why use assymetrical conditioning with more caps on negative rail?
Much obliged if someone can help me out here, trying to understand best practices for module design.
New Here : EOE

euromorcego
Super Deluxe Wiggler
Posts: 2757
Joined: Sat Jan 18, 2014 6:22 am

Re: Eurorack (?) Module Power Filter / Conditioning Capacito

Post by euromorcego » Mon Nov 19, 2018 2:29 am

maniac303 wrote: Question is....why use several small (100nf) caps in parallel, (followed by a larger (10 uF) polarized cap)? Why not just use a single bigger cap, followed by the even bigger polarized cap? Is there an advantage to using a handful of different devices? Is there something else going on here that I don’t understand? For example see Befaco modules where they use 10 or 11 caps where 2 would do?
because schematics do not represent physical location of components. The string of 100n caps you see in schematics is distributed across the board, they should be close to the power pins of the IC.

User avatar
cygmu
Super Deluxe Wiggler
Posts: 1507
Joined: Mon Feb 15, 2016 3:33 pm
Location: UK

Post by cygmu » Mon Nov 19, 2018 4:27 am

Just to add: the physical location matters because in real life PCB traces, solder connections, component leads etc all have resistance and inductance, so where you place your caps makes a difference. Capacitors have some resistance too, and so on.

The datasheets for most ICs will give some clues about what kind of power supply decoupling is recommended. For split-supply devices the usual recommendation is to add decoupling caps between each supply rail and ground.

User avatar
maniac303
Learning to Wiggle
Posts: 14
Joined: Sun Oct 14, 2018 9:35 pm
Location: Vancouver

Thanks for your answers!

Post by maniac303 » Mon Nov 19, 2018 10:13 am

Thanks folks, your answers make perfect sense. That's what I get for studying schematics and not spending enough time looking at boards. :doh:
New Here : EOE

User avatar
mskala
Super Deluxe Wiggler
Posts: 2614
Joined: Thu Mar 27, 2014 7:33 am
Location: Toronto
Contact:

Re: Eurorack (?) Module Power Filter / Conditioning Capacito

Post by mskala » Mon Nov 19, 2018 10:40 am

maniac303 wrote:Question: Some module builders/designers condition the provided 12v power further, right on the module, with zener diode followed by ferrite bead, followed by a string of parallel caps.
You probably mean a Schottky diode; Zeners have their place in voltage regulation but aren't what you usually see used for reverse protection at module power inputs.

As well as the points others mentioned, there's the issue that not all types of capacitors are the same. The "polarized" capacitors you see are normally aluminum electrolytics, which don't behave well at high frequencies. They're used to filter out stuff like power supply ripple, which is low-frequency and relatively high energy, at the scale of the entire board. But the high-frequency spikes created for instance by some digital circuits will not be well-filtered by an electrolytic; the inherent resistance and even inductance of the big capacitor prevents that large capacitance from being available to absorb the spike. For high-frequency stuff you need a ceramic cap very close to the specific IC. So you'll often see large caps used in parallel with small caps even though naive electrical theory would suggest that just making the large cap a tiny bit larger and leaving out the small one, ought to be equivalent.

User avatar
maniac303
Learning to Wiggle
Posts: 14
Joined: Sun Oct 14, 2018 9:35 pm
Location: Vancouver

Re: Eurorack (?) Module Power Filter / Conditioning Capacito

Post by maniac303 » Mon Nov 19, 2018 1:54 pm

mskala wrote:You probably mean a Schottky diode; Zeners have their place in voltage regulation but aren't what you usually see used for reverse protection at module power inputs.
Yes, correct, my mistake.
mskala wrote:As well as the points others mentioned, there's the issue that not all types of capacitors are the same. The "polarized" capacitors you see are normally aluminum electrolytics, which don't behave well at high frequencies. They're used to filter out stuff like power supply ripple, which is low-frequency and relatively high energy, at the scale of the entire board. But the high-frequency spikes created for instance by some digital circuits will not be well-filtered by an electrolytic; the inherent resistance and even inductance of the big capacitor prevents that large capacitance from being available to absorb the spike. For high-frequency stuff you need a ceramic cap very close to the specific IC. So you'll often see large caps used in parallel with small caps even though naive electrical theory would suggest that just making the large cap a tiny bit larger and leaving out the small one, ought to be equivalent.
I understand the need for different sizes based on what you describe above, and am used to seeing two caps with different sizes per rail for that reason, but was curious about the use of several smaller caps with the same value. Now that I understand the topology it makes perfect sense. I was wondering if there was some 'magic' in using several in parallel vs. one larger one that I failed to appreciate, better thermal disapation for example... but that was before considering the topology and the need to have them adjacent to the power pins on the ICs to clean the high frequency stuff, some of which could be generated in situ on the board. I should have figured it out by studying the boards... my bad.

Thanks again all! :hail:
New Here : EOE

KSS
Super Deluxe Wiggler
Posts: 2677
Joined: Mon Jan 25, 2016 7:28 am

Re: Eurorack (?) Module Power Filter / Conditioning Capacito

Post by KSS » Mon Nov 19, 2018 10:43 pm

maniac303 wrote:I was wondering if there was some 'magic' in using several in parallel vs. one larger one that I failed to appreciate, better thermal disapation for example...
Not directly related to your first question, one additional point is that ESR of a series of caps goes down according to Ohms law. You see this in PSU filter caps, for example. Where there may be several equal size smaller caps adding up to the larger value desired ripple requires. Helps with smaller case size too.

So there is some available 'magic' using several in parallel, it just doesn't apply to your original question.

User avatar
Drilldoughzer
Wiggling with Experience
Posts: 269
Joined: Sun Jun 11, 2017 12:34 am
Location: La belle province
Contact:

Post by Drilldoughzer » Sun Sep 08, 2019 5:13 pm

Quick noob question:

If I want to filter the +5 rail;

1-what electro cap should I use?

2-The + of the cap to +5 and - to ground?

Thanx!

Post Reply

Return to “Music Tech DIY”