Quieting noisy grounds?

[Lance Morgan]

I'm seeing 0.15 V noise in both the power and sensor ECU grounds (on a
scope), in an older Saab 900, which may be contributing to a 0.25 V
Air Mass Meter variance at a static RPM (and/or there may be other
contributing causes), i.e., if the throttle is held fixed at 2800 RPM,
the AMM voltage fluctuates (a misfire of sorts, and an audible engine
misbreathing/pulsing problem). Trying to discern between an ignition
or fuel misfire, has been unyielding so far (can provide more info).
AMM voltage stabilizes/normalizes at higher static RPMs

What maximum ground voltage would be cause for concern... anything
more than 0 V? Is this essentially a ground spike? Or is it too low a
V?

Using 12 AWG wire, I've jumpered the ECU grounds directly to the
battery neg terminal, w/o change.

Could a diode or MOSFET be safely used? if so, what in particular?


http://68.18.136.52/electrical_system_folder/lh24wiringnfmp.htm

ECU = 200

Intermittent 0.15 V, at various time bases, seen at power grnd ECU pin
17 and sensor grnd pin 5, to engine head 201

AMM = 205

0.25 V variance seen at AMM signal output pin #3 and ECU #7 (at static
2800 RPM), unknown if related to ECU ground noise

Thank you,
Lance

 

[Guy Macon]

Lance Morgan <alancemor@yahoo.com> says...

I'm seeing 0.15 V noise in both the power and sensor ECU grounds (on a
scope),

A voltage is, by definition, between two points. When you say
that you have 0.15 volts of noise, it raises the question
"referenced to what?" In other words, when you attached the
tip of your scope probe to your ground, what did you attach
the ground of your scope to?


--
Guy Macon, Electronics Engineer & Project Manager for hire.
Remember Doc Brown from the _Back to the Future_ movies? Do you
have an "impossible" engineering project that only someone like
Doc Brown can solve? My resume is at http://www.guymacon.com/

 

[CFoley1064]

Subject: Quieting noisy grounds?
From: alancemor@yahoo.com (Lance Morgan)
Date: 4/5/2004 9:14 PM Central Standard Time
Message-id: <2a52f3e9.0404051814.4b351e82@posting.google.com

I'm seeing 0.15 V noise in both the power and sensor ECU grounds (on a
scope), in an older Saab 900, which may be contributing to a 0.25 V
Air Mass Meter variance at a static RPM (and/or there may be other
contributing causes), i.e., if the throttle is held fixed at 2800 RPM,
the AMM voltage fluctuates (a misfire of sorts, and an audible engine
misbreathing/pulsing problem). Trying to discern between an ignition
or fuel misfire, has been unyielding so far (can provide more info).
AMM voltage stabilizes/normalizes at higher static RPMs

What maximum ground voltage would be cause for concern... anything
more than 0 V? Is this essentially a ground spike? Or is it too low a
V?

Using 12 AWG wire, I've jumpered the ECU grounds directly to the
battery neg terminal, w/o change.

Could a diode or MOSFET be safely used? if so, what in particular?


http://68.18.136.52/electrical_system_folder/lh24wiringnfmp.htm

ECU = 200

Intermittent 0.15 V, at various time bases, seen at power grnd ECU pin
17 and sensor grnd pin 5, to engine head 201

AMM = 205

0.25 V variance seen at AMM signal output pin #3 and ECU #7 (at static
2800 RPM), unknown if related to ECU ground noise

Thank you,
Lance

It sounds like you're heading in the right direction. In older cars, you
frequently have a resistive connection between the battery and the chassis.
There is a ground strap between the battery and chassis. Try redressing this
connection, using a wire brush to get a clean metal connection. If you have
any doubt, drill and tap a new hole for the ground strap bolt. If you have any
doubt about the integrity of the crimp connection, replace the whole ground
strap. This has helped me so many times that I now do it pre-emptively
whenever I get a used car (I prefer older ones, too -- they're more
economical).

By the way, you're posting in the wrong group here. You might want to try
rec.autos.tech for better results. Many engineers and technicians here work on
their own cars from time to time, but you'll have more luck in that group.

Good luck
Chris

 

[Lance Morgan]

Guy Macon <http://www.guymacon.com> wrote in message news:<zLqdnQUEq5atre_d4p2dnA@speakeasy.net>...

Lance Morgan <alancemor@yahoo.com> says...

I'm seeing 0.15 V noise in both the power and sensor ECU grounds (on a
scope),

A voltage is, by definition, between two points. When you say
that you have 0.15 volts of noise, it raises the question
"referenced to what?" In other words, when you attached the
tip of your scope probe to your ground, what did you attach
the ground of your scope to?

I adjust the channel's initial vertical DC trace position, to coincide
with the Ground ref line (switching back and forth between DC and
ground)

The probe head is contacted to the test point(s). The probe's ground
clip is, via various short routes (some require extension), attached
to the battery negative post. I've also grounded the probe to various
chassis points.

The scope itself is grounded thru the AC three prong outlet, grounded
to earth. I have not made use of the front panel's chassis ground
connection (should I?)

I've tried to follow the techniques described at

http://www.tek.com/Measurement/App_Notes/ABCsProbes/60W_6053_8.pdf

the section starting at Fig 6.1 & thru Advanced Probing Techniques (I
use a non-mainstream PDF reader or would offer the page #).

I've tried various connection paths, and probably don't have enuf
experience to recognize extraneous noise vs true signal noise, and/or
ringing or abberation.

 

[Lance Morgan]

cfoley1064@aol.com (CFoley1064) wrote in message news:<20040406020926.04141.00000676@mb-m17.aol.com>...

Subject: Quieting noisy grounds?
From: alancemor@yahoo.com (Lance Morgan)
Date: 4/5/2004 9:14 PM Central Standard Time
Message-id: <2a52f3e9.0404051814.4b351e82@posting.google.com

I'm seeing 0.15 V noise in both the power and sensor ECU grounds (on a
scope), in an older Saab 900, which may be contributing to a 0.25 V
Air Mass Meter variance at a static RPM (and/or there may be other
contributing causes), i.e., if the throttle is held fixed at 2800 RPM,
the AMM voltage fluctuates (a misfire of sorts, and an audible engine
misbreathing/pulsing problem). Trying to discern between an ignition
or fuel misfire, has been unyielding so far (can provide more info).
AMM voltage stabilizes/normalizes at higher static RPMs

What maximum ground voltage would be cause for concern... anything
more than 0 V? Is this essentially a ground spike? Or is it too low a
V?

Using 12 AWG wire, I've jumpered the ECU grounds directly to the
battery neg terminal, w/o change.

Could a diode or MOSFET be safely used? if so, what in particular?


http://68.18.136.52/electrical_system_folder/lh24wiringnfmp.htm

ECU = 200

Intermittent 0.15 V, at various time bases, seen at power grnd ECU pin
17 and sensor grnd pin 5, to engine head 201

AMM = 205

0.25 V variance seen at AMM signal output pin #3 and ECU #7 (at static
2800 RPM), unknown if related to ECU ground noise

Thank you,
Lance

It sounds like you're heading in the right direction. In older cars, you
frequently have a resistive connection between the battery and the chassis.
There is a ground strap between the battery and chassis. Try redressing this
connection, using a wire brush to get a clean metal connection. If you have
any doubt, drill and tap a new hole for the ground strap bolt. If you have any
doubt about the integrity of the crimp connection, replace the whole ground
strap. This has helped me so many times that I now do it pre-emptively
whenever I get a used car (I prefer older ones, too -- they're more
economical).

By the way, you're posting in the wrong group here. You might want to try
rec.autos.tech for better results. Many engineers and technicians here work on
their own cars from time to time, but you'll have more luck in that group.

Good luck
Chris

Thanks for the response. All of the above has been done, to no avail.
In the past, I've sometimes had luck posting this type of topic to
this NG, and also alt.engineering.electrical. There are certainly
knowledgeable electrical guru's on r.a.t, but they don't always
frequent the group. Likely this query it too basic for s.e.d, but
hopefully a reasonable/acceptable cross-post

Lance

 

[Guy Macon]

Lance Morgan <alancemor@yahoo.com> says...

Guy Macon <http://www.guymacon.com> wrote...

Lance Morgan <alancemor@yahoo.com> says...

I'm seeing 0.15 V noise in both the power and sensor ECU grounds (on a
scope),

A voltage is, by definition, between two points. When you say
that you have 0.15 volts of noise, it raises the question
"referenced to what?" In other words, when you attached the
tip of your scope probe to your ground, what did you attach
the ground of your scope to?

The probe head is contacted to the test point(s). The probe's ground
clip is, via various short routes (some require extension), attached
to the battery negative post. I've also grounded the probe to various
chassis points.

I understand. Here is something that you can rule out with a simple
test; the loop formed by the scope's ground clip to the battery post
back to the ground point you are measuring is an antenna that may
pick up extraneous noise. The way to rule this out is to keep the
scope's ground clip as it is and touch the scope tip to the battery
negative post. Your actual noise will disappear but any noise that
is picked up by the loop will remain.

The scope itself is grounded thru the AC three prong outlet, grounded
to earth. I have not made use of the front panel's chassis ground
connection (should I?)

The scope being grounded through the AC three prong outlet is fine.
To insure a solid measurement, make sure that this is the only
connection between the vehicle and earth. You can test this with
an ohmmeter. Vehicle tires are surprisingly conductive for something
that looks like rubber (lots of carbon it them) but garage floors
are often insulators as long as they are dry. If needed, one can
always put cardboard between tire and floor.

I've tried to follow the techniques described at

http://www.tek.com/Measurement/App_Notes/ABCsProbes/60W_6053_8.pdf

You just did me a big favor by accident; I have been looking for
the "ECB toprobe tip adaptor" shown in figure 6-2, page 37, and
nobody in sci.electronics.design knew where to buy one from my
description. Now I have a picture! But I digress...

the section starting at Fig 6.1 & thru Advanced Probing Techniques (I
use a non-mainstream PDF reader or would offer the page #).

Page 36.

I've tried various connection paths, and probably don't have enuf
experience to recognize extraneous noise vs true signal noise, and/or
ringing or abberation.

Ringing isn't an issue for you. As you can see from figure 6-1, ringing
only happens when there is a fast step change. You are measuring a
ground trace and don't care if any step changes ring a bit.

Now that I understand the situation a bit better, let's revisit your
original post and see if I can be of assistance...

--------------------------------------------------------------------

Lance Morgan <alancemor@yahoo.com> says...

I'm seeing 0.15 V noise in both the power and sensor ECU grounds (on a
scope), in an older Saab 900, which may be contributing to a 0.25 V
Air Mass Meter variance at a static RPM (and/or there may be other
contributing causes), i.e., if the throttle is held fixed at 2800 RPM,
the AMM voltage fluctuates (a misfire of sorts, and an audible engine
misbreathing/pulsing problem). Trying to discern between an ignition
or fuel misfire, has been unyielding so far (can provide more info).
AMM voltage stabilizes/normalizes at higher static RPMs

This gives us a chance to partially isolate grounding issues. When
you go to higher RPMs and the AMM voltage stabilizes/normalizes, is
the 0.15 V noise you are seeing in the power and sensor ECU grounds
still there? If so, I would suspect that it's not the cause of the
AMM voltage fluctuations at lower RPMs. Check it with the scope,
though; it could be that the AMM voltage is still fluctuating but
the engine isn't sensitive to the noise at higher RPMs.

What maximum ground voltage would be cause for concern... anything
more than 0 V? Is this essentially a ground spike? Or is it too low a
V?

If it weren't for you reported misbreathing/pulsing problem, I
wouldn't worry about this at all. Try this; do the same test on
any sensor ground on any other vehicle you have laying about;
I suspect that they will have much the same measurement.

Using 12 AWG wire, I've jumpered the ECU grounds directly to the
battery neg terminal, w/o change.

That tends to make me think it's not a ground noise problem,
but I am not 100% sure.

Could a diode or MOSFET be safely used? if so, what in particular?

I doubt if it will help, but it's really easy to try, so you might
as well give it a shot. Get a 1N4001 (or 1N4002 to 1N4007, they
will all work the same in your application) and connect the anode
(bottom of the triangle on the schematic) to ground and the cathode
(bar on the schematic) to 12V. Do this right at the sensor, keeping
the leads fairly short. Vehicles tend to have really nasty negative
spikes that are hard to see on a scope, and he diode will suppress
them. The reason I think that this won't help is because the sensor
was designed to ignore the kind of spikes often seen in automobile
electrical systems. It's cheap and easy to try, though.

http://68.18.136.52/electrical_system_folder/lh24wiringnfmp.htm

ECU = 200

Intermittent 0.15 V, at various time bases, seen at power grnd ECU pin
17 and sensor grnd pin 5, to engine head 201

AMM = 205

0.25 V variance seen at AMM signal output pin #3 and ECU #7 (at static
2800 RPM), unknown if related to ECU ground noise

Also see:

http://68.18.136.52/electrical_system_folder/lh24text.htm

Let's try this:

Put your scope ground lead on pin #1 of 205 and measure all of
the pins (#1 through #5) with the probe tip. Tell me what you see
on each pin.





--
Guy Macon, Electronics Engineer & Project Manager for hire.
Remember Doc Brown from the _Back to the Future_ movies? Do you
have an "impossible" engineering project that only someone like
Doc Brown can solve? My resume is at http://www.guymacon.com/

 

[w_tom]

Noise between two points that should be same ground? To see
it, learn and use oscilloscope in differential mode. Noise
measurements are only valid between two point - which is two
scope probes are required. Currently you are looking at
common mode 'noise'. Noise that may not even exist from the
perspective of electronics. To see noise that electronics
sees, do differential mode observations.

Lance Morgan wrote:

I'm seeing 0.15 V noise in both the power and sensor ECU grounds (on a
scope), in an older Saab 900, which may be contributing to a 0.25 V
Air Mass Meter variance at a static RPM (and/or there may be other
contributing causes), i.e., if the throttle is held fixed at 2800 RPM,
the AMM voltage fluctuates (a misfire of sorts, and an audible engine
misbreathing/pulsing problem). Trying to discern between an ignition
or fuel misfire, has been unyielding so far (can provide more info).
AMM voltage stabilizes/normalizes at higher static RPMs

What maximum ground voltage would be cause for concern... anything
more than 0 V? Is this essentially a ground spike? Or is it too low a
V?

Using 12 AWG wire, I've jumpered the ECU grounds directly to the
battery neg terminal, w/o change.

Could a diode or MOSFET be safely used? if so, what in particular?

http://68.18.136.52/electrical_system_folder/lh24wiringnfmp.htm

ECU = 200

Intermittent 0.15 V, at various time bases, seen at power grnd ECU pin
17 and sensor grnd pin 5, to engine head 201

AMM = 205

0.25 V variance seen at AMM signal output pin #3 and ECU #7 (at static
2800 RPM), unknown if related to ECU ground noise

Thank you,
Lance

 

[Guy Macon]

w_tom <w_tom1@hotmail.com> says...

Noise between two points that should be same ground? To see
it, learn and use oscilloscope in differential mode. Noise
measurements are only valid between two point - which is two
scope probes are required. Currently you are looking at
common mode 'noise'. Noise that may not even exist from the
perspective of electronics. To see noise that electronics
sees, do differential mode observations.

Given the skill level of the person asking the question (skilled
in automotive repair, unskilled in electronics) and the high
probability that the "scope" we are talking about is an automotive
scope as opposed to an electronics engineering scope, (think
Sun/Snapon rather than Tektronix) I wouldn't advise differential
measurements. Connecting the scope ground to the vehicle ground
at the ground wire of the sensor will, in my opinion, work just
fine.


--
Guy Macon, Electronics Engineer & Product Manager for hire.
Remember Doc Brown from the _Back to the Future_ movies? Do you
have an "impossible" engineering project that only someone like
Doc Brown can solve? My resume is at http://www.guymacon.com/

 

[w_tom]

0.15 volt noise can be quite normal when using a scope in
common mode. IOW the person is encouraged to read the manual
farther since the person wants to go beyond simple mechanical
repair. Ground noise of 0.15 volts can be seen on scope in
common mode BUT may not exist from perspective of sensors.
The better skilled learn and use differential mode to observe
what the sensor input sees. Raise the skill level if tracing
down 0.15 volt noise. Learn to use that scope in differential
mode - to see what the noise voltage is between two points.

Guy Macon wrote:

Given the skill level of the person asking the question (skilled
in automotive repair, unskilled in electronics) and the high
probability that the "scope" we are talking about is an automotive
scope as opposed to an electronics engineering scope, (think
Sun/Snapon rather than Tektronix) I wouldn't advise differential
measurements. Connecting the scope ground to the vehicle ground
at the ground wire of the sensor will, in my opinion, work just
fine.