>Dan Michaels wrote:
>>
>> Tony Nixon wrote:
>>
>> >
>> >You're right, I only catered for about 470uF at present, but the PCB is
>> >still 'open' for awhile yet. I figured what I learnt at electronics
>> >school - 1000uF / amp (ball park).
>
>Terrible advice me thinks.
>
>> I/C = dv/dt --> 1A/1000uF = 1000v/sec
>> --> 1v/msec --> 20v/20msec, where 20msec = period of 50hz AC
>>
>> [Check my math - it's too early in the morning]. It says, if you suck
>> 1A out of a 1000uF cap over the period of your AC input, then Vcap
>> will drop 20v. Yes?
>
>My heads a bit heavy from last night, but here goes..
>
>12VAC 1A wall wart unloaded gives around 13.8VAC
>
>The PicPocket draws 50mA quiescent and 100mA while programming.
>
>Wall wart output with a 130mA load is around 13.6VAC
>
>13.6VAC = 19.2VACPeak
>
>1.2V drop through bridge = 18VAC peak
>
>I need 13V + 2V min for 78xx regulation.
>
>18 - 15 = 3V
>
>Take a reasonable(?) margin and say 1V ripple allowed.
>
>C = Iload / 2 X F X DV
>
>C = .1 / 2 X 50 X 1 (60Hz for US even better)
>
>C = 1000uF
>
>Take a narrow margin and allow 2V ripple and I can use the 470uF.
>
>Probably better with the 1000uF though, in case the ICSP circuit draws a
>bit of current.
>
>For about $2 more, could use LM2930-8 instead of 7808 for better ripple
>margin.
>
>--
>Best regards
>
>Tony
>
>mICro's
>
http://www.picnpoke.com
>
EraseMEsalespicnpoke.com
>
>--
>
http://www.piclist.com hint: To leave the PICList
>
RemoveMEpiclist-unsubscribe-requestEraseMEEraseMEmitvma.mit.edu
>
>
>
>
>
>