Hi all!
As I'm mostly not putting a battery jack into my monstrosities, I was wondering what's the bare minimum of DC power filtering that gets the job done? Right now I'm thinking a 100R in series and a 100uF to ground, but as with anything I'm more than willing to be taken to school?
Thanks!
I think you've got it. I'd still add the polarity diode. Just in case. Should also help with static discharge on IC's
1N4001 in parallel, positive to ground?
As a bare minimum for power filtering, just a 47u cap fits the bill.
Better is a LPF made from a 100R or 47R resistor and 100u or 47u cap.
Even better is the above plus an extra 100n ceramic cap in parallel with the larger cap.
Thanks CJ! I think that's the way I'm gonna go and forget about polarity protection. Most of what I make never leaves my house anyways. :-)
sent from my mobile device
Well, I would say that power filtering and power supply protection are two different things.
The typical 1N4001 in parallel with the power rails does the job, but it's far from perfect. I prefer a schottky diode in series with the positive rail. This provides the same reverse-polarity protection as the other method, but it won't cause either the DC wall wart or the diode to die.
Quote from: culturejam on February 26, 2014, 05:52:48 PM
Well, I would say that power filtering and power supply protection are two different things.
The typical 1N4001 in parallel with the power rails does the job, but it's far from perfect. I prefer a schottky diode in series with the positive rail. This provides the same reverse-polarity protection as the other method, but it won't cause either the DC wall wart or the diode to die.
I agree completely. I use a 1N5817 in series with the power input in everything now.
Quote from: culturejam on February 26, 2014, 05:52:48 PM
The typical 1N4001 in parallel with the power rails does the job, but it's far from perfect. I prefer a schottky diode in series with the positive rail. This provides the same reverse-polarity protection as the other method, but it won't cause either the DC wall wart or the diode to die.
This makes up for the 1590A hipsters crack in the other thread.
Quote from: midwayfair on February 26, 2014, 06:57:09 PM
This makes up for the 1590A hipsters crack in the other thread.
Thanks. I was doing the series diode thing waaaaaaay before everybody else jumped on the bandwagon. 8) ;D
Quote from: midwayfair on February 26, 2014, 06:57:09 PM
Quote from: culturejam on February 26, 2014, 05:52:48 PM
The typical 1N4001 in parallel with the power rails does the job, but it's far from perfect. I prefer a schottky diode in series with the positive rail. This provides the same reverse-polarity protection as the other method, but it won't cause either the DC wall wart or the diode to die.
This makes up for the 1590A hipsters crack in the other thread.
Wait, Forrest was bagging on 1590a builds???
/ offended
// not really
Thanks all! Much obliged!
So if I go with a Schottky, is 100R necessary?
Quote from: muddyfox on February 27, 2014, 09:19:16 AM
So if I go with a Schottky, is 100R necessary?
The diode is for polarity protection, the resistor and cap are for filtering.
OK, I was just wondering, since diode is bound to have some sort of resistance to it?
How about this one from R.G.? Is it strictly polarity protection?
http://www.geofex.com/article_folders/cheapgoodprot.htm
There is some resistance in the diode, but it's almost always vanishingly small (a few ohms). But anyway, having the resistor there is cheap and effctive, and it doesn't take up much space. You could even do it with a SMD resistor to save more space. :)
OK to sum it up, first I put a 100R and a 1n5817 and then after that I put a 100uF and a 100nF in parallel to ground?
Quote from: muddyfox on February 27, 2014, 10:16:28 PM
OK to sum it up, first I put a 100R and a 1n5817 and then after that I put a 100uF and a 100nF in parallel to ground?
I think you need to be a bit careful with the value of the resistor you use. All of the current that the effect use will flow through that resistor. So, an effect that draws a lot of current will drop a lot of voltage over that resistor.
For effects that are op-amp and IC based, you might be better off not using the input resistor and putting 100nF ceramic caps on the power pins of all the IC's.
I see. Since there's apparently no generic solution that's good enough for all cases, I need to start writing this down. ;D
Thanks!