Wow, thank you for the headache so early in the morning! I got my one tube circuit working, but going to a much larger uC, and more tubes, more noise is bound to come up, so now that I know how to decouple with caps, I'll be sure to add them. Thank you all. Such a great group. I hope you all know how sincerely I appreciate the help. I'm in love with glowing glass, and I'm in love with arcane technology. Without you guys, I'd still be trying to figure out power supplies, and binary decoding.
Shane On Thu, Jul 21, 2011 at 9:14 AM, jb-electronics <webmas...@jb-electronics.de > wrote: > ** > Hi Shane, > > > When you say "next to" and IC, do you mean from the +5V, to ground? > > > Yep. There are some IC sockets who have a 100nF capacitor connected from > the pin on the bottom left to the pin on the top right, i.e. the most common > IC power pins. > > > Resistors I got, Capacitors frustrate me... > > > It is not that hard to understand: Capacitor act just like resistors for > AC. The higher the frequency, the more current can flow through a capacitor. > The complex impedance is Z := 2*Pi*(-i)/f*C, where f is frequency in Hertz, > C is the capacity in Farad and i is the imaginary basis, e.g. i^2 = -1. This > just means that the impedance (the resistance, basically) approaches zero if > the frequency approaches infinity. > > In our example: Really fast disturbances can be seen as some very high > frequency (look at the Fourier transform of the signal). These disturbances > will be shorted by the small capacitor of 100nF due to their high frequency, > so that this high frequency does not corrupt your circuit. > > Slow signals are not affected. In the other limes, the frequency is zero > (i.e. DC signals), so the impedance is infinite. We already know that: > Capacitors do not conduct DC current. > > Hope this helps. > > Best regards, > Jens > > > > On Thu, Jul 21, 2011 at 4:16 AM, Nick <n...@desmith.net> wrote: > >> Decoupling should be done EVERY time on EVERY chip - analogue or >> digital - right adjacent to the supply pins. >> >> It costs pretty much nothing, and saves a whole bunch of trouble. Just >> do it. >> >> Part of the need to do this is that chips today are much much faster >> than they used to be, so where frequency response would roll off >> before oscillation, nowadays even standard opamps can have GBWs in the >> MHz to 10s of MHz range, and logic goes far far higher. >> >> On the analogue side, I'm currently restoring some Quad amps - the >> amount of pure twaddle on the www about using loony opamps like the >> OPA627 and much faster (in "audiophoolery" faster = better) class A >> drivers & output stages - recipe for high-frequency instability - the >> circuit were designed to use the inherent limits of the original >> devices. >> >> Maybe I'll just spend 1000 bucks on some speaker cables and >> unidirectional 99.99999% OFC internal cabling. Not. >> >> Nick >> >> On Jul 21, 9:50 am, jb-electronics <webmas...@jb-electronics.de> >> wrote: >> > Hi, >> > >> > > (2) Sprinkle capacitors across power and ground all over your circuit. >> > > Preferably as close to the power and ground pins of each chip as >> > > possible. Usually they're 0.1uf (100nf) ceramic capacitors. Some big >> > > chips require you to use several near them, so read you datasheets. >> > > Chips are fast. Very fast. They can either generate very brief short >> > > circuits (in the ballpark of 10nS), and/or be susceptible to these >> > > very short glitches on the power rails. >> > >> > I cannot stress enough how much pain this will spare you. I recently >> > built a combined volt- and amperemeter with a 2x16 LCD readout on a >> > rather small pcb, and I did point-to-point-wiring like I always do, and >> > it did not want to work. Some weird oscillations at the volts ADC. The >> > first thing I did was inserting a 100nF cap next to every (!) IC, and >> > bam, problem solved. >> > >> > Jens >> >> -- >> You received this message because you are subscribed to the Google Groups >> "neonixie-l" group. >> To post to this group, send an email to neonixie-l@googlegroups.com. >> To unsubscribe from this group, send email to >> neonixie-l+unsubscr...@googlegroups.com. >> For more options, visit this group at >> http://groups.google.com/group/neonixie-l?hl=en-GB. >> >> > -- > You received this message because you are subscribed to the Google Groups > "neonixie-l" group. > To post to this group, send an email to neonixie-l@googlegroups.com. > To unsubscribe from this group, send email to > neonixie-l+unsubscr...@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/neonixie-l?hl=en-GB. > > > -- > You received this message because you are subscribed to the Google Groups > "neonixie-l" group. > To post to this group, send an email to neonixie-l@googlegroups.com. > To unsubscribe from this group, send email to > neonixie-l+unsubscr...@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/neonixie-l?hl=en-GB. > -- You received this message because you are subscribed to the Google Groups "neonixie-l" group. To post to this group, send an email to neonixie-l@googlegroups.com. To unsubscribe from this group, send email to neonixie-l+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/neonixie-l?hl=en-GB.
