From: sademade (sademade_at_yahoo.com)
Date: 2001-12-22 05:30:14
> So the pcbs were not really a cost factor until now.
Just wanted to save you from exposure to toxic chemicals, plus
increase the shielding potential of the boards, if using 2 layers
and copper pours etc. And of course, making it easier for the rest
who would purchase copies of the boards ... But if you think this is
a bad idea, of course, everyone can continue to etch boards home in
the kitchen sink, that makes wife/girlfriend so happy.
>> I have used BBrown ADS1211 4 diff channel A/D part in some
>>projects,
> A nice chip (with an awful reset option that I have never got
>working).
Got it working. If one carefully calculates the timing sequence,
and applies it, it works, sometimes, it requires a few rounds
of the sequence with a do while c-loop, but it works each time
within some milliseconds, no problem. Have a c source for it, if
you ever would want to compare your routines with my simple-as
ones, which appear to work, though.
> I have used it in a project too recently (16 channel pH controller)
Ok. Teflon stands appeared to be a must with pH applications, or
did you find board material to get by without need for air-wiring
or teflon standing ?
> Maybe it is possible to get below 1uVp-p noise.
If that would be possible, it would be great !
> However a bit costly (at least at RS Components here in Germany).
Not very costly here, and makes life easier, for those of us who
are not the best with analog design, and expect the same parts
would in guru hands produce superior results ... after all, whether
the units costs $20-30 more in parts, does not make much difference,
if it helps it to be smaller and better.
> These isolators would have a much higher data rate than the RS232
>line I suppose ?
Yes, I have used isolators that handle up to 10 megabits fine,
but have limited the rate to 5 megabits so far, do not need
more with the AD. They are dual-SO-8 parts from HP. Original
reason to get those was CAN-buses 1 megabit rate, and wanted
to have some "margin". Just so easy to use, and requires much
less drive from CPU pins, so one can do away with additional
drive transistor or gate that would take more space, and reduces
the need for pull-ups and such, producing smallest space design.
> Moritz and I have had big trouble to get a CNY64 (which is rated
>for IEC601-1)working at 115200 baud.
> So I see some problems...
Sure. CNY's are beatifully cheap, but, they are slow as hell.
The HP parts are awesome, and there are also some quads from
Infineon, that seem to perform faster, and yet, do not cost that
badly for a quad, and they lay out well, too, very small SO-10
pack.
> I think these are good ideas. However at the moment
> we (or at least I )should focus on getting the modularEEG
> analog part working.
Definitely, and agree 100 % with that. Partially, the AD-idea
was thrown to see if it could help making the analog part working,
trough possibly removing the need for the second stage amplifiers,
and therefore, possibly helping to reduce noise, and thinking was
because of not being good analog designer, would have tried that
as a cowards way of using few buck more expensive parts to get
rid of some problems, to overcome that not all of us are as good
analog designers as we should, but, if you succeed in the cheaper
design, the better, and if anyone on the list, like Jim, who are
good in analog, can assist you to get the analog done without
need for more powerful A/D's that would be great for all of us.
Does anyone have CircuitMaker = source SPICE models for the INA114,
would like to try to simulate the INA114 front stage with the CM
as do not have MSIM8, though latter is surely better.
If one sims, how could one sim the noise pick-ups, ie, how to
find simming tools that could even roughly estimate the noise
trouble, and allow one to design something less susceptible to
noise, do any sims help with shielding/active shield driver
designs ?
Thank You, Sade M.
This archive was generated by hypermail 2.1.4 : 2002-07-27 12:28:35 BST