home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!spool.mu.edu!agate!ames!sun-barr!news2me.EBay.Sun.COM!exodus.Eng.Sun.COM!crashnburn.Eng.Sun.COM!rmp
- From: rmp@crashnburn.Eng.Sun.COM (Richard Pottorff)
- Newsgroups: rec.audio
- Subject: Re: quiet current sources. Any ideas?
- Date: 23 Dec 1992 18:05:49 GMT
- Organization: Sun
- Lines: 55
- Distribution: usa
- Message-ID: <ljhajtINN53v@exodus.Eng.Sun.COM>
- References: <1992Dec18.203205.20576@porthos.cc.bellcore.com> <Fyg3VB1w165w@kunikpok.UUCP> <1992Dec21.150028.1331@porthos.cc.bellcore.com>
- NNTP-Posting-Host: crashnburn
-
- In article <1992Dec21.150028.1331@porthos.cc.bellcore.com> gfl@herahera.cc.bellcore.com (lenahan,grant f) writes:
- rg@kunikpok.UUCP (R.G. Keen) writes:
- > gfl@herahera.cc.bellcore.com (lenahan,grant f) writes:
- >
- > > I need to modify an input buffer for lower noise.
- > >
- > > Can anyone suggest a method for making a _quiet_
- > > 15-ish mA constant current source? The olde zener-
- > > bipolar solution is proving noisy.
- > >
- >
- > Try bypassing the zener.
-
- Oh, I've tried quite a few things. Understand that a B-E
- junction is very noisy, and that the effective impedance of a
- current source is very high. Thus voltage noise across it
- is also high.
-
- Combine this with the amplifying effect of a BJT on any
- voltage reference noise, and the problem becomes significant -
- and hard to fix.
-
- The answer will fall in one of two areas:
- 1 - filter the BE junction (can't filter output or zener)
- 2 - find an alternate current source
-
- I prefer the latter, and still solicit inputs. But about 20 hours
- of various people's time have already gone into this - its become
- annoying!
-
- Thanks again, and thanks for your suggestion. Regrettable its
- not so simple as a noisy Zener.
-
- Grant
- gfl@hera.cc.bellcore.com
-
- I suggest:
-
- Why not use a large resistor (in relation to your impedance) in series with
- a appropriately valued voltage source?
-
- E=IR. R=E/I. Let E=1, then R=1/15mA; R=66.67 Ohms. OK, that's not a big
- enough resistor. Lets scale E and thus R by using a 20V supply.
- R=1333.333... Ohms.
-
- What happens, is the size of R swamps any changes of resistance (if its
- quite a bit (at least an order of magnitude) smaller than R. You can keep
- pushing R up if you keep pushing E up as well.
-
- Rick
- --
- Stolen signature:
- GILLETTE'S PRINCIPLE:
- "If you want to make people angry, lie. If you want to make
- them absolutely livid with rage, tell the truth."
-