[sdiy] Singin

Fri Mar 14 17:47:01 CET 2003

Aye matey, but it be worse than that...

The LM324 has a PNP input stage that will always
try to source its bias current (now don't jump
me on what we call current source and current sink)

If there is not a DC path for this bias current... it
will charge your S/H cap to the positive rail in no time...

well, in real time.... Make the cap big and it charges slower.

I've always said the LM324 sucks... this is just one more
evidence of same.

(well it IS cheap and works good for a lot of single supply
apps - I do have them and use them)

H^) harry

--------Original Message--------  

From: madhun2001 at yahoo.com
To: Synth-Diy <synth-diy at dropmix.xs4all.nl>
Sent: Mar 14 2003 01:05
Subject: [sdiy] Singin

>Yo,
>Play me a durge, matie ... I be tellin' a tale of
>sample and holds.
>Yes, the drama is required. Silly me, I thought an
>LM324 would be okay for a sample and hold buffer. It
>looked good on the scope. Sadly, ears are very
>sensitive to tiny little changes in pitch. So when my
>okay-looking s&h output was tied to my VCO, well, it
>was not good.
>A quick look-see at the datasheets tells the tale: The
>LM324 has worse-case 100nA of input bias current. The
>TL074 is about 500 times better at 200pA input bias
>current. Luckily, the pinout is the same.
>So tomorrow I will try it and see if it fixes the
>problem.
>Sigh...
>--Tim
>__________________________________________________
>Do you Yahoo!?
>Yahoo! Web Hosting - establish your business online
>http://webhosting.yahoo.com