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: firstname.lastname@example.org (c3l)
Subject: Re: Digital design help needed.
Date: 24 Sep 2002 05:58:21 -0700
NNTP-Posting-Date: 24 Sep 2002 12:58:21 GMT
JHB@jita.nospam.demon.co.uk (Jim Backus) wrote in message news:...
> On Mon, 23 Sep 2002 19:25:31, email@example.com (c3l) wrote:
> > The problem I face is the noise generated by the Schmitt trigger. It
> > seems (from the oscilloscope) that when the schmitt triggers the data
> > acquisition board, the is a slight spike in the output of the S/H
> > amplifier.
> It's a long time since looking at sample and hold specs, but I
> remember that sample feedthrough has always been a problem. In other
> words it's more likely to be a limitation of the S/H than the schmitt
> trigger. Good layout practices - separate analogue and digital grounds
> with a carefully chosen single interconnection, keeping the sampling
> signal well away from the analogue signal etc will all help but
> ultimately I think there will always be some feedthrough from the
> sampling to the sampled signal.
I use a power supply with two different supply output i.e. +-12V and a
+-5V and I have two separate ground for the S/H amp ( one signal
ground and one supply ground) and both these ground are connected only
at one point on the pcb and that point is then connected to the +-12V
On the other side of the pcb board(same board), the Schmitt and
Monostable share the same supply and signal ground which is then
connected to the +-5V.
Both these sides ( the S/H side and the Schmitt/Monostable) are
separated and not connected on the PCB board. Basically there is an
empty space on the PCB with no metalisation except for one track from
the Monostable to the the S/H which carries the HOLD signal.
The suppply and signal ground of the S/H amplifier is only connected
to the supply and signal ground of the Schmitt and Monostable through
the power supply.
Physically, the sampling signal is about 5cm away from the sampled
Any comments on my ground plane design?
I couldnt find the feedthrough data on the datasheet but I could find
the hold mode feedback and it gives a -88dB for a 20Vpk-pk sinusoidal
How do I calculate the amount of feedtrough I'm getting from a 20us
pulse width sampling signal with a 5V amplitude? There isnt a graph on
the datasheet which shows the variation of the hold feedback with
Thanks for your help. I will try a different layout and see if I get
any improvements. The funny part about this is, I did a initial design
on a breadboard, and both boards i.e. the PCB and the breadboard gave
This really worries me! ( my digital design is really bad )
Another observation I've noticed about the noise spike. It only
appears on the S/H amp output when I probe it with the oscilloscope or
when its connected to the data acquisition board. When the schmitt
output is left floating, the S/H amp output is perfert i.e. no noise
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup