ping

[Carter Shore]

I'm always amazed that automotive ECM can work
reliably with such high impedance sensors. The
underhood environment is very hostile and noisy. I
have a hardware and software background, with some
aeronautical and communications thrown in. Noise
management was always a problem.
I worked on a computer once that was located in the
offices of a TV station. They were getting
intermittent memory parity errors. I put a scope on
the +5V buss, and was seeing 3V p-p of RF modulation!
We added a fistful of bypass capacitors and shielded
the case to solve the problem.

Maybe we should be using BNC connectors and such.

--- Jeff M <tystorm at email.msn.com> wrote:
> I had an interesting time working on a friends Turbo
> Grand Prix.  It came to
> me with a rough idle so, slapped on the Diacom and
> found a consistent line
> of 7 degrees knock retard all the time, cool, a new
> one on me (just wait).
> No engine noises and all things are fairly new on
> this car (I put in a while
> ago; new injectors, some sensors and such), so I
> swap out the ECM with a
> known good one (that is easier than swapping the
> knock sensor) and still 7
> degrees all the time.  Swap out the knock sensor and
> still.  Pull out the
> (single) knock wire from both connector ends and ohm
> it out, good wire and
> not going to ground, even when shaking the engine by
> hand.  Check volts out
> of ECM, fine here.  Pull out wire and visually
> inspect its entire length,
> good again, check wire loom and no pinch or running
> near anything that could
> cause a false ?  Tired of this stuff I just grab
> some twisted pair/shielded
> cable and run a new wire in a new wire loom and in a
> different location and
> presto, 7 degree knock gone and been so for 8 months
> (also tested it still
> work by the usual thump test, and some higher boost,
> as we tuned up his chip
> for 100 octane).  I also took the braided shield of
> this wire and on one end
> (did not want to create a ground loop) secured it to
> a good ground, with
> other ground wires.
> 
> (always up to a new challenge)
> Jeff Middaugh
> tystorm at email.msn.com
> 
> 
> > You need to watch the timing with some type of
> scan tool.  I would do this
> > before spending money on anything else.
> > Mark  By the way, I know why you would twist the
> wires.  I have just never
> > heard of a problem with rfi on the knock sensor
> circuit.  The ecm is set
> up
> > to look for specific frequency so I doubt rfi
> would trigger knock retard.
> > Mark
> > -----Original Message-----
> > From: David Mulvey <kd6bzn at QNET.COM>
> > To: gmecm at efi332.eng.ohio-state.edu
> <gmecm at efi332.eng.ohio-state.edu>
> > Date: Monday, August 23, 1999 11:36 PM
> > Subject: Re: ping
> >
> >
> >
> >
> > > It sounds to me like the knock sensor is NOT
> working and/or egr problem.
> > The
> > > best way to check the knock sensor is to take a
> plunger handle, and drop
> > it
> > > with it's own weight only onto the block near
> the knock sensor.  Watch
> the
> > > timing, (Scan tool) and make sure the ecm is
> pulling some timing out.  I
> > > have never heard of twisting the wires.  I have
> ALL the tech bulletins
> on
> > > the L98.  I have tuned the knock sensor a bit by
> using a bit of teflon
> > tape
> >
> > The twisting of the wires is just a simple way of
> keeping some low
> frequency
> > RF from resonating on a stran of wire.  As long as
> the
> > wire twists the RF has only very small untwisted
> sections to resonate.
> >
> > I checked the EGR valve and the EGR solenoid and
> the voltage to the
> > solenoid.  All Ok, a good vaccuum to the egr valve
> caused the
> > motor to idle down..  I cleaned out the EGR
> solenoid gave it a test on a
> 12
> > volt power supply. So EGR I think it fine.
> >
> > I do have one old injector on the side of the
> engine that doesn't have the
> > O2 sensor. Could be a suspect.  I have 3 old
> injectors on
> > the O2 sensor side, and do not get any mixture
> errors.
> >
> > I was wondering about the Knock sensor too.  I do
> get error 43 which I
> think
> > is the code showing that the knock count exceeded
> it's
> > limits. An audible ping/knock is also heard at
> this time.  I guess I was
> > wondering if the knock sensor maybe was only
> working a little
> > not detecting the all of the knocks then causing
> the computer not to
> adjust
> > properly then setting off the MAF voltage too high
> and/or
> > too low..  Kinda getting complicated at this
> point.
> >
> > I still need to check the ground to the MAF
> sensor.
> > I still need to give the knock sensor wires a
> couple of twists.
> > I want to replace the 4 old injectors I have,
> (short on funds at the
> moment)
> > AutoZone has injectors for $35 each.
> > Am I buying crap or are these things workable?
> > AutoZone also has MAF sensors for $139,
> > $80 core charge. (got one the other day, same
> 'ping')
> > practically everything else is new except for the
> ESC module.
> > I figured the ESC is probably working since
> > I do have a car that runs.. Good assumtion?
> > I need to see if the knock sensor is working using
> the plunger stick test.
> > I haven't gotten into changing ECM values yet. 
> Will not hesitate if
> nothing
> > else seems to work.
> > "PE vs. RPM table"  =- What's 'PE'?
> >
> > For the Knock sensor test would I notice a timing
> change using the plunger
> > handle and a timing light? I got one of those..
> >
> >
> > end
> > ----
> > David Mulvey KD6BZN
> > Lancaster, CA
> > (661) 723-0352
> >
> >
> 
> 
> 
> 
> 
> 

__________________________________________________
Do You Yahoo!?
Bid and sell for free at http://auctions.yahoo.com