From: sleeper75se (sleeper75se_at_yahoo.se)
Date: 2002-02-01 01:46:27
--- In buildcheapeeg_at_yahoogroups.com, "Jim Meissner" <jpmeissner_at_mindspring.com> wrote:
>The RF from the radio station and equipment switching glitches are
> not attenuated. The output impedance of the op amp is not low
> enough at these frequencies. In your case you even have a 220 ohm
> resistor in series to make matters even worse. The shields would
> actually drive noise into your groundplane under the input
> resistors and right into the input stage.
Actually the ground-plane under the resistors is connected to the
shield, not ground. On both sides. Bad idea?
> I am not impressed with simulations done on a computer.
> The real world does not care what you simulated.
Actually, I haven't simulated the shield driver. It's just touch and
go. :-)
Ok, how about this solution: The 100pF capacitors connect to the
shield instead. Then the shield driver's primary job would be to keep
LF-noise (such as 50/60Hz hum) out, while the caps take care of the
HF.
> If I were to use your board, and I certainly want to, I would first
> rip out all the driven shield components and tie J100 pin 3 to
> ground. C100 and C101 are already on ground. The rest looks good.
> (BTW I would not use the filter section either)
I could design a layout without shield driver and filter if you
like...? It is not much work, I'll just rip out whatever is not
needed.
>
> Could you make these little input boards first and test them? That
is by far the most important step.
> After that everything is much less critical.
I'm on it.
--- In buildcheapeeg_at_yahoogroups.com, "Jim Meissner" <jpmeissner_at_mindspring.com> wrote:
> I looked at the A/D converter section and I think it looks good
> enough the way it is to build and test.
>
> Full speed ahead. The testing will tell the story.
I just got the order confirmation from Digikey. It really is full
speed ahead. They obviously didn't notice that I wanted FedEx
International Economy, they ship it with FedEx International Priority
(2 days) instead. Shipping is *only* 48 bucks (notice the sarcasm).
Thing is, my thesis is due in a week, so I'll have to slow down the
EEG project for a while...ho hum.
Regards,
Andreas
This archive was generated by hypermail 2.1.4 : 2002-07-27 12:28:37 BST