[sdiy] Re: Bypass caps

[Seb Francis]

Recently when I was trying to draw too much current from a 723 based PSU 
(in my Oakley 3031) I saw a sawtooth type waveform on the power rail 
(not going anywhere near 0V, just ramping over a small voltage range) - 
but this was when drawing only just a bit more current than the set 
limit, so perhaps when much more current is drawn the effect will be 
different.  But then again the 723 PSU in question does not have any 
kind of latching-off mechanism - it will simply keep trying and trying 
to up the voltage, hit the current limit, lower the voltage, etc, etc.

Seb


Harry Bissell Jr wrote:

> No it does not. I'm using the standard "PowerOne" type supply...
> if it gets an overcurrent, it shuts down the output and remains latched
> off until power is reset.
>
> If you power off, then on again quickly, it will start because the caps
> are not 'all the way' discharged and the current spike is less...
>
> that is still ugly...
>
> H^) harry
>
> */Seb Francis <seb at burnit.co.uk>/* wrote:
>
>     surely having a current limiting PSU (e.g. with an LM 723) would just
>     bring the voltage up as the caps charged?
>
>     seb
>
>
>     Harry Bissell Jr wrote:
>
>     > Just for the record (and noobs)
>     >
>     > another solution is to either
>     >
>     > Power one half the rack, then after a time delay
>     > of a second or so... power the other half using a relay.
>     >
>     > ~or~
>     >
>     > Power the whole system through some series resistors that limit the
>     > inrush current, then short across those resistors after a one second
>     > delay.
>     >
>     > This is how I did my system... but the 'brute force' larger
>     supply is
>     > a valid option
>     > too. Its more size and weight, less complexity.
>     >
>     > Engineering... ~make the choice~
>     >
>     > H^) harry
>     >
>     > */Tim Daugard /* wrote:
>     >
>     > From: "harrybissell"
>     > .
>     > >
>     > What price failure ? More time is wasted by NOT using sufficient
>     > bypass caps than is saved by not designing them in (in small
>     > production runs). It is really hard to OVERDO the caps, unless the
>     > power supply is unable to start because of the peak current :^P
>     > >
>     >
>     > Which happens, my system reached the point where the power
>     supply was
>     > adequete for steady stae use, but couldn't power up the system.
>     I had
>     > to unplug one row of modules from the power chain until the system
>     > powered up. When all the other modules had charge their caps,
>     the last
>     > chain got pluged in. I finally solved that problem by putting in an
>     > upgraded power supply - two modules wide with 3X the current
>     > capability.
>     >
>     > Tim Daugard
>     > AG4GZ 30.4078N 86.6227W Alt: 12 feet above MSL
>     > http://home.sprintmail.com/~daugard/synth.htm
>     >
>     >
>     >
>     >
>
>