EMI Problems?

[Steve Gorkowski]

On the data logger side of things use shielded twisted pair wire ground only one
side . On race car stuff I have went to 4-20 ma current loop to keep things clean
. For power I use a sealed lead acid battery for power source and the pc power
supply that runs off a 12v battery then use a pc104 single board computer in a box
that has a ground strap ( box on rubber mounts ) with rf feed thru on power wires
going into box.

Andrew Wakeling wrote:

> Hi All,
>
> I'm not sure if this topic's been discussed previously before, but I'm
> going to start it off nevertheless. OK, EMI (Electromagnetic
> Interference) is a popular topic discussed throughout the electronic
> world and it is known to cause many strange problems with electrical
> devices. Devices had to pass certain EMI tests and they had standards...
> CE marks might ring a bell. Anyhow, what does this all have to do with
> the Delco you ask?
> I have a strange feeling that many problems on high load/high rpm may be
> caused by EMI produced by the spark lead or ignition modules. A friend
> of mine had such a problem and moved his coil pack around the car, with
> suprising results. The car basically went from running badly, to not
> running at all.
> What I also discovered was that if spark leads were changed, and the
> standard ones were replaced, the ECM was more prone to these problems. I
> am also experiencing similiar problems with my data logging device and
> think that its due to this EMI interference... however, I am unsure how
> Diacom gets around it.
>
> Dave Zug wrote:
>
> > my website (under the buick GN section) has a link to pictures of the graphs
> > during this glitch - high load, high RPM, it just goes berzerk - and the car
> > goes into 31 degrees retard!  It does cause I felt the power loss big time.
>
> This problem I think is associated with EMI. Diacom may keep on logging,
> and maybe its interface, being composed of simple transistor networks
> and the like, are less prone to EMI. Since Diacom is also being used in
> the parallel port, a serial routine isn't being used, making
> interference less significant maybe? If Diacom was designed so that it
> would actually follow the whole serial transmission instead of just
> points along the byte, maybe results would be 100% accurate.
>
> Does anybody have any views on this? Has anybody solved any problems of
> the such with insulation or shielding? Anybody know how Delco filters
> signals without shielding their cables? Is Elvis still alive? If you
> have any answers, please reply to the group!
>
> Andrew Wakeling.