On Sun, 4 Apr 1999, Donald Gaffney wrote:
> I'm impressed that you can run a coherent closed loop at 70kHz. At this
> rate it seems like the jitter would, at least in the limit, be on par with
> the task period.
Keep in mind this is a hardware triggered closed loop, not an RTL periodic
task - we use one of the onboard counters on the DAQ board to generate the
periodic interrupt. We did timings using both a scope as well as the other
onboard 20MHz counter on the DAQ board (with the timing triggered within the
interrupt routine, so as to consider any software/latency effects). We
averaged over 2000 cycles, on a given timing trial, worse case jitter was
never worse than 2 microseconds (and typically around 1 us) (this is from
memory, so I could be wrong -- lab notebook is in another lab right now).
I hope to have a WWW page up sometime this summer describing our application
as well as the software. The only think stopping me from doing it right now
is that I am changing jobs/academic affiliations and (as is always the case)
have a zillion things to do to make my coworkers happy before I move next
month.
Which leads me to a shameless plug:
Any students on here? Graduate _or_ undergraduate. I am looking for someone
who would be interested in continuing development of my project, which
involves an RTL application used for real-time control of neurobiological
experiments (for those who are familiar with neurophysiology, it is an
open-source project to implement the "dynamic clamp" technique, but runs a
lot faster than previously published methods). Tasks to be accomplished
include:
- porting the code to use COMEDI
- overall improved modularization/compartmentalization of the code
- writing a COMEDI driver for an unsupported DAQ card
- development of a GUI and and a GUI communication mechanism
(ideally in LabView, so that the GUI can run on a variety of platforms,
not just Linux. It is hard enough convincing a neuroscientist to even
put _one_ Linux box in the lab ...)
- development of some sort of parser/compiler than can take a simple
description of a control algorithm input by a user and create a
real-time control module that implements the users algorithm.
This could be for class credit, or as part of an advanced degree.
As far as location, I will be in the ECE Department at Georgia Tech in
Atlanta, GA. Students in both ECE and Biomedical Engineering are OK. The
lab where the experiments are currently done is at the National Institutes
of Health in Bethesda, Maryland. My coworkers there said they would be
willing to host a student who is enrolled at any of the nearby engineering
schools in and around Washington, DC.
Email me for details. Thanks!
--
Rob Butera, Postdoctoral Fellow http://mrb.niddk.nih.gov/butera/
Laboratory for Neural Control, NINDS
National Institutes of Health, Bethesda, MD USA
--- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
----
For more information on Real-Time Linux see:
http://www.rtlinux.org/~rtlinux/
