The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: "Ian Buckner"
Subject: Re: generating calibration voltage
Date: Wed, 8 Jan 2003 10:44:53 -0000
Organization: Agilent Technologies
NNTP-Posting-Date: Wed, 8 Jan 2003 10:44:57 +0000 (UTC)
X-Newsreader: Microsoft Outlook Express 6.00.2600.0000
X-Cache: nntpcache 2.3.3 (see http://www.nntpcache.org/)
"Tom Torfs" wrote in message
> Hello John,
> > How square must the square wave be? (What rise and fall times do
> > require?)
> Rise/fall times can be upto a few ms. I'm only interested in the
> amplitude. The reason I can't just use a DC signal is that there is
> highpass filter with fc=0.1Hz after that (and also a lowpass filter
> with fc=70Hz).
> > and how much capacitance to ground can you tolerate with
> > this floating output?
> Something like 10 pF would be ideal.
> > What precision must the amplitude hold to?
> Preferrably something on the order of magnitude of a few uV.
> > How flat must the wave be (can there be any ringing on the flat
> > of the wave?
> I would think the lowpass filter will take care of that, but with
> high risetimes allowed I don't think there's any reason for it to
> To the people suggesting a transformer: that's very nice, but I do
> have room for a 30Hz transformer. The footprint of the entire design
> is limited to a few square cm... that's why I was hoping for some
> suggestions of doing this with some sort of opamp circuit.
> My alternative if a floating output turns out to be too difficult is
> to briefly hook up the reference to ground during calibration using
> switch with low impedance when on and very high impedance when off.
> could also use a differential mux per channel, but rather avoid that
> as that would add many components (24 channels).
I think you are asking for an accuracy of a few tenths of 1%.
If so, start with a precision reference voltage, e.g. the LT1021 at
Use a precision attenuator from the likes of Vishay or Caddock to get
your 1mV output, then use the LTC1043 to a) float the output, then
b) chop it at your required rate to produce your square wave.
If you are concerned about thermoelectic effects, do the float/chop at
5V/10V level, then precision attenuate (probably adds 1 extra
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup