Electrical Grounding

[KenH]

One of my winter projects is to improve the grounding on the car sensors to help with a stumble issue I have with the VEC2. While doing a little research on the topic, I see that the ricer crowd is getting into special grounding kits that improve overall system grounding and even seem to gain some HP during independent magazine testing. One example can be seen here:
http://www.caraudiomag.com/testreports/0401cae_hyper/

Looking at the Viper ground system, it looks like some significant improvements can be made given the rear battery mounting location and the fact that the system grounds have to run the length of the steel frame.

My thinking was to use high-end stereo ground wiring to run a heavy gauge cable from the battery up to the engine compartment and then use a power distribution block to connect smaller gauge wire to key components and sensors.

I'm curious if anyone has tried something similar.

 

[ruckdr]

KenH,
Be sure to Dyno before and after you make the changes, and under the SAME CONDITIONS!!!!!!
They say they gained 13.7 RWHP at 5700 RPM, thats 7.3%.
If I were to get 7.3% increase, I would get 30.4 MORE RWHP.
Quite frankly - I ain't buying those kinds of RWHP increases.
My intuitive opinion (no facts).
Later,

 

[joe117]

I read the article in the link you posted.

I can tell you this, IT IS PURE 100% BUNK!!

There is no science to anything they say.
There is no more horsepower to be gained by special grounding.

The rear battery, with it's negative terminal connected to the rear of the frame, is grounded just as well as any other grounding position.

Pay close attention here,
This grounding nonsense is coming from the same people who would have you believe that "special" speaker cables work better than simply good speaker cables.

These folks are known for their pseudo science when it comes to grounds and cables.
They write articles talking about how cables have to be "broken in" before they work properly.

They believe that cables have some kind of memory and thus cant be turned end for end.

They believe that various size conductors in an audio cable will make it work better.

They ignore the real science and make up the most ridiculous theories in order to convince non technical people to spend huge $$ for NO IMPROVEMENT. Not even a little.

I sure hope you aren't one of those audio or home theater guys who had bought into the expensive cable fraud.

This grounding system is nothing but a way to get your money. PERIOD.

And if you make your own special grounding system, you will still get nothing in the way of improvement.
It will just cost less.

 

[ruckdr]

Thanks Joe117,
I think we would all be jumping for something like this if the HP gains were as they implied - see above,
Later,

 

[joe117]

What's funny is that it's from a car audio magazine.
The audio people, car or otherwise, are always coming up with bs.
Lots of non technical people believe it though. The audio people talk about cables and grounds as if real science didn't exist.

 

[KenH]

I don't care about audio quality, just about the integrity of the grounding of the engine sensors, etc. From discussions with Sean Roe, it appears that some of the issues that people have run into have been related to poor sensor grounding, especially on the later GenII cars. While preparing to simply ground the sensors to the frame, it occurred to me that there may be a better way to go about it and if I am dorking with it anyway, I may as well do it right.

For the naysayers, I wouldn't be so quick to dismiss the idea that there may be something to be gained by improving the overall grounding system on the Viper whether it be through smoother operation or some additional HP. Forgetting audio quality for a minute which wasn't even part of my question, conductor size, conductor strand size, conductor length and conductor material all have an affect on impedance and how good of a ground is presented to an electrical signal. The steel Viper frame with the battery hooked to one end and the motor and sensors hooked to the other end is hardly an ideal conductor by anyones standard, but perhaps it is adequate.

 

[Tom F&L GoR]

When I started work long ago at a big fat dinosaur oil company, one of the projects I worked on could be called the grandfather of today's GDI engines. It was a multi-fuel engine that would run on any fuel that could be pumped. Part of the system was an alternating current ignition system. The ignition trigger was a pressure sensor on the direct injection fuel line; the idea was that for as long as fuel was injected (like a diesel, perhaps 30 crank angle degrees,) the ignition would create an AC arc across the plug gap.

The engine would run on engine oil, gasoline, alcohol, etc, and an OEM wanted to look at it on an emissions dyno. We sent it to Detroit, but they couldn't get the engine to run beyond a stumbling idle. The AC ignition system needed Viper sized ground straps to satisfy the current flow. So, young engineer crimps loop ends on wire braid, attaches to ignition box and to block, fixes problem, is (temporary) hero.

Anyway, perhaps nothing to do with the original question, but a trip back in time.

 

[joe117]

You say,
"conductor length and conductor material all have an affect on impedance and how good of a ground is presented to an electrical signal. The steel Viper frame with the battery hooked to one end and the motor and sensors hooked to the other end is hardly an ideal conductor by anyones standard,"

Ok,
You tell me the science.

1. What is the impedance of the Viper frame in respect to the ground when the battery is grounded to the rear?
What is the change in impedance when it is connected to the front?

2. What would cause the "smoother operation" or "additional HP" in respect to special grounding?

3. What would the conductor strand size have to do with the grounding of a sensor?

4. You state that "a battery hooked to one end and the motor and sensors hooked to the other end is hardly an ideal conductor by anyones standard"
Why do you think this is true? Exactly what does it do to the signals you are talking about?

5. What is the frequency and voltage level of the sensor signals you are talking about?

If a sensor has problems because of "poor sensor grounding", it is because there is no wire connecting the sensor to the frame or perhaps a gasket or sealant is in the way. It's not because the wire isn't the right kind or size.
Either way, it's not something that needs improvement.
It just needs to be fixed.

If anyone thinks there is a grounding issue, they would simply connect a voltmeter between the sensor body and the frame of the car.
If you read zero volts, IT IS GROUNDED.

I can see that you have bought into this "special grounding" crap.
And,
from what you hinted about "Forgetting audio quality for a minute", I suspect that you probably believe at least some of the audio cable hokus pokus that the audio industry charlatans spit out to confuse non technical people.

If you can even define the term "impedance" I would be surprised.

Please don't tell people that some special grounding system will get them an improvement over the properly functioning stock grounding connections, unless you can back it up with science.

I don't believe you can come up with anything.
I guess I'm calling bs on you.

 

[Jerry Dobson]

I am a huge fan of grounding the motor at multiple locations. It never hurts, it can only help.

I don’t know about getting any extra HP. From skimming the article, it looks like they did one dyno run, added ground, second dyno run, and then claimed more HP. Huh? What about removal of the ground and a third dyno run? I think the second run would have shown an increase with no changes.

I am somewhat concerned about adding ground within the ECU/PCM boundaries. Most of these systems are “Isolated Grounds” and not direct grounds. I have heard nothing but good things about adding the grounds as ROE suggests, I am just a little concerned without looking at the schematics to see what I would be bypassing. I think one of the really early posts on this had a comment from one of the Dodge engineers that said … No No.

 

[Steve 00RT/10]

Without going into a full explanation, there is a black with blue striped
wire coming from the PCM (where you tapped the a/f meter brown wire).
This is a ground from the PCM that goes to 12 major PCM sensors (MAP
sensor, TPS, IAT, Oxygen sensors, etc). This ground wire has tremendous
resistance, causing the sensors to send the wrong signal to the PCM. We
have found that by splicing into this wire at 1 or 2 points boosts the
ground and significantly improves the running of the engine (smoother).
It has also knocked out the O2 heater codes on the cars tested .......

This info is courtesy of Sean Roe who I would guess has as much experience with the PCM and electrical engine management of the Viper as any other. I grounded a couple more of these sensors last May...........just for GP. What I believe he is saying is that the 12 sensors are getting their return to the PCM over one #20 or #22 AWG wire. I think when you're talking tenths of a volt to a signal, there is not enough voltage push to overcome the resistance in that small a wire with 12 sensors all using it at the same time. Therefore an inaccurate signal can sometimes be relayed to the PCM which then could produce runnability issues. His testing seems to bear this out.

Steve

 

[joe117]

So,
"12 sensors are getting their return to the PCM over one #20 or #22 AWG wire. I think when you're talking tenths of a volt to a signal, there is not enough voltage push to overcome the resistance in that small a wire with 12 sensors all using it at the same time.

Why do you say that?

Do you just feel it in your bones?

Do you have any idea how much current is involved?

Do you know how much resistance there is in 20 feet of #20 wire?

 

[Ulysses]

standard copper wire specs:

20 AWG 10.4 OHMs / 1000 ft. @ 77 Deg F
22 AWG 16.5 OHMs / 1000 ft. @ 77 Deg F

Not to be taken as the word of God since manufacturers tend to vary by an OHM / 1000 ft.

 

[Steve 00RT/10]

I am only re-phrasing the information given to me by Sean Roe. I, as most, do consider him an authority in this area. The quote I used above was from him. I guess you have issues with him, not me.

I see the resistance values for #20 AWG have been posted. I would have guessed an ohm or less in 20'. If we assume .2 volt at 1 ohm of resistance...then jogging my memory fom the mid 70's....the current would be .2 amps. I don't begin to profess to know how all these sensors work in conjunction with each other, but 12 sensors continually sending signals at .2 volts would be 2.4 amps. I would guess this much juice would heat up a #20 or #22 wire. Heat is resistance. Therefore, more resistance could cause some false voltage readings when they the reached the brain. I could be wrong, but that's how I interpreted what he told me..............Other than that----it's definitely in the bones.

Steve

 

[KenH]

1. What is the impedance of the Viper frame in respect to the ground when the battery is grounded to the rear?

/** Steel has a higher impedence than copper. Ohms law E=IR says that if my return path has a higher resistance (R) and the same current (I) then I will lose voltage (E) meaning that the ground leg of the sensor will rise above ground potential and thereby offset the analog output that the sensor is providing. This will also allow the ground leg to act as an antenna and collect noise impulses from sources like the spark plug wiring. **/

What is the change in impedance when it is connected to the front?
/**It will inherently be less since you have a shorter path through a relatively poor conductor such as steel. **/

2. What would cause the "smoother operation" or "additional HP" in respect to special grounding?

/** See comment above. Sensors will tend to read more accurately and with less spurious noise impulses. That can only help in both these cases. Whether it is enough to make a noticeable difference would depend on how good the stock system is. **/

3. What would the conductor strand size have to do with the grounding of a sensor?

/**Primarily through the 'skin' effect and how it relates to higher frequency 'noise' on the ground wire and it's ability to shunt it to ground. It is the reason that braided shielded cabling provides better performance than foil shielded cabling **/

4. You state that "a battery hooked to one end and the motor and sensors hooked to the other end is hardly an ideal conductor by anyones standard"
Why do you think this is true? Exactly what does it do to the signals you are talking about?

/** Covered above. The chassis is made of steel, a relatively poor conductor in the scheme of electronics. There is a reason your house power is not wired with steel wiring. **/

5. What is the frequency and voltage level of the sensor signals you are talking about?
/** From 0 to 8 volts or so and 0hz to 100hz I would guess. **/

If a sensor has problems because of "poor sensor grounding", it is because there is no wire connecting the sensor to the frame or perhaps a gasket or sealant is in the way. It's not because the wire isn't the right kind or size.
Either way, it's not something that needs improvement.
It just needs to be fixed.
/** way over simplified. You have obviously never worked in a product design environment. If you think everything you buy has been 100% optimized for performance from the factory, you're not living in the real world **/

If anyone thinks there is a grounding issue, they would simply connect a voltmeter between the sensor body and the frame of the car.
If you read zero volts, IT IS GROUNDED.
/** So if I hook my meter up to the chassis and coffee cup and I read zero volts, you're telling me that my coffee cup is grounded? I don't think so. Lack of a difference in potential between two objects does not gurantee that they are grounded together, though it can be an indication. Your also making the assumption that the sensor body has to be grounded and that may not be part of the requirement for the sensor in question **/

I can see that you have bought into this "special grounding" crap.
And,
from what you hinted about "Forgetting audio quality for a minute", I suspect that you probably believe at least some of the audio cable hokus pokus that the audio industry charlatans spit out to confuse non technical people.
/** Who are you calling non-technical people? **/

If you can even define the term "impedance" I would be surprised.
/** 25 years in electronics tells me I probably can. What's your background on this topic? **/

Please don't tell people that some special grounding system will get them an improvement over the properly functioning stock grounding connections, unless you can back it up with science.

I don't believe you can come up with anything.
I guess I'm calling bs on you.
/** Put your reading glasses on and re-read this topic. At no time have I made this statement. I only stated that improvements have been noted elsewhere and wondered if anyone had any actual experience with this as I was thinking about trying it myself. I'll chalk you up as opinionated, but with no actual experience to back up your opinion. Does that about cover it? **/

 

[Ulysses]

I'm going to try to expound on what Jerry Dobson said, because I believe he hit it right on the head. I also believe that attaching those sensor grounds to frame ground is bad idea.

Most of these systems are “Isolated Grounds” and not direct grounds. I have heard nothing but good things about adding the grounds as ROE suggests, I am just a little concerned without looking at the schematics to see what I would be bypassing. I think one of the really early posts on this had a comment from one of the Dodge engineers that said … No No.

The circuit in the PCM will most definitely have it's own ground plane or "isolated" ground from which the signals/powers to the sensors are generated. These signals/powers will be referenced to that internal ground plane which will be at a different potential from the frame ground. Also the circuitry involved in the deciphering and decision making process will be using this internal ground as a reference for the return signals from the sensor.

In other words, a signal sent out as 5V will be 5V as referenced to it's internal ground. By moving to frame ground you have changed/offset the voltage from the sensors since frame ground will be at a lower potential than the internal ground reference.

Let's say for the sake of argument that the frame ground is at 0V but the internal ground is at .1V. A signal is generated at 5V referenced to .1V and sent to the TPS. The TPS in turn returns a 2.5V signal. That signal, when the sensor ground is moved to the frame, is now offset by .1V.

EDIT:

Check it out, I found this in the shop manual.

SENSOR RETURN-PCM INPUT

OPERATION
The sensor return circuit provides a low electrical noise ground reference for all of the systems sensors. The sensor return circuit connects to internal ground circuits within the Powertrain Control Module (PCM).

 

[Steamroller]

According to an online service manual database, all of the grounds (that is, the sensor grounds and power grounds) are internally connected inside the PCM, which would indicate they all provide the same reference (that is, they are not isolated from each other). However, it is further indicated that the sensor grounds have noise suppression, so they are uniquely different from the power grounds.

As Sean Roe indicated, splicing in additional ground lines appeared to help solve the issue. According to the information in the above paragraph, doing what Sean did could potentially introduce noise into the sensor's signal, as ground paths without noise suppression are being provided. Of course, what he did appears to have worked, so maybe any additional noise introduced was insignificant. Thus, his efforts could be indicating that signal level is more of an issue than signal noise.

Has anyone considered splicing in larger sensor wires? I realize there would be a notable amount of labor involved, but it would keep the original wiring architecture intact.

 

[KWIK96]

This story was writen by a good salesman.
How many ground wires do they run on F1 cars, or any car for that matter?
One.

The ricer crowd will snap them up in their never ending search for more HP
(you gotta feel sorry for them)

I won't be rushing out to purchase something which will only add extra weight to my car, and as an added bonus, make it look damn ugly.

Time to ground this thread. LOL

 

[KenH]

Finally, a little productive conversation on the subject.

Ground is a relative thing. I have worked on aviation electronics where ground was a 300VAC waveform relative to earth ground. I also agree that ground relative to what the PCM considers ground rather than chassis ground is what is important as that is how the PCM will interpret the incoming signal. The PCM ground does find its way down to the chassis ground lug above the side sill and eventually finds its way back to the battery by running the length of the chassis frame.

Noise suppression on the ground return lines (and signal lines) at the PCM is a common technique. Generally that consists of ferrite beads or similar which help to scrub some of the noise off the ground line before it can get back into (or out of) the electronics. This is generally done for EMI and protection of the electronics rather than for improving sensor reading accuracy per se. We deal with designing electronics and sensors that reside near high voltage arc lamps that have an ignition pulse of 5KV to 25KV, not unlike a spark plug wire. It is surprising how much noise can get coupled into the sensor lines causing bad sensor readings or even damage to the controller electronics.

I believe that the crux of the problem is that the ground as seen at the sensor is not the same ground as the ground referenced by the PCM.

The K4 ground circuit which is the sensor grounding circuit is composed of 18AWG wire which goes to about a dozen sensors from the front air temp sensor all the way back to the back battery temp sensor before it connects to a single pin on the PCM. I would estimate that this wire would have to be at least 30' long with all its stubs if not a lot more. Being so long, relatively thin, being designed with a number of stubs out to the sensors and crossing so many areas of the car, that cable is going to act a bit like an antenna and will pick up some noise from the other electrical systems in the car. The map sensor is very close to the ignition wires for instance and could easily have some noise coupling into the ground wire. The sensor grounds do not appear to be grounded at any of the sensors as someone incorrectly noted above. I am sure that under static conditions, the grounds near the sensors would all read close to gnd (<1 ohm) with a DVM. Under dynamic conditions with the motor running I am sure that the ground has a fair amount of noise on it if checked with a scope.

The sensor outputs back to the PCM are referenced to the ground at the end of this wire where it is connected to the particular sensor, not at the PCM end where all the wires converge on the PCM ground. For instance, the TPS is a potentiometer referenced to the ground wire on one side and +5V on the other side. If the ground wire at the sensor end where to bounce up by .5V due to a noise spike on the line, the output of the TPS would change by about 10% relative to the more solid ground as referenced back on the PCM.

Grounding the sensor ground to the chassis at some of the distal points, while not providing an optimal ground is probably still better than the ground as seen on the cable and hence the sensor and closer to the ground as seen at the PCM, leading to the improvement that has been seen by some.

My thinking to improve upon this (without rewiring the whole ground system) was simply to add a heavy gauge copper cable from the battery ground (the best ground on the car) up to a point near the front of the car and use it to establish a more solid ground than the chassis can provide for both the PCM and the sensors. Star wiring from the newly established ground point out to some of the key sensors would just be tapped into the existing ground wires at the sensors.

If I get some time, I will drag a digital storage O'scope home and take some snapshots of the ground/signal performance at the sensors and perhaps bring a little science into the discussion. Unfortunately, the most interesting phenomenon may not show up while idling in the garage. If the issue can be observed and quantified, the effect of any grounding changes can be better quantified.

I'm just guessing here, but I'm expecting about a 50HP SOTP improvement from the change

 

[Ulysses]

If the ground wire at the sensor end where to bounce up by .5V due to a noise spike on the line, the output of the TPS would change by about 10% relative to the more solid ground as referenced back on the PCM.

I don't think you would get a ground bounce without a power bounce as well that is in phase.

I do think that 30' of wire is a lot of wire. And a xmas tree layout is definitely not optimal.

 

[joe117]

1. What is the impedance of the Viper frame in respect to the ground when the battery is grounded to the rear?

/** Steel has a higher impedence than copper. Ohms law E=IR says that if my return path has a higher resistance (R) and the same current (I) then I will lose voltage (E) meaning that the ground leg of the sensor will rise above ground potential and thereby offset the analog output that the sensor is providing. This will also allow the ground leg to act as an antenna and collect noise impulses from sources like the spark plug wiring. **/

The massive steel frame when viewed as a conductor is large enough for its resistance to be ignored. Measure it and see. Calculate the difference between a frame made of steel and what it would be if it were copper. You will find no real world difference.
When a conductor is very large, it can't be improved by making it very, very large.
---------------------------------
Also, the "impedance" and the "resistance" are the same for the conductors and frequencies we are dealing with.
Non technical audio people often are confused by this. Right?
---------------------------------
3. What would the conductor strand size have to do with the grounding of a sensor?

/**Primarily through the 'skin' effect and how it relates to higher frequency 'noise' on the ground wire and it's ability to shunt it to ground. It is the reason that braided shielded cabling provides better performance than foil shielded cabling **/

Of course you know that skin effect has no effect when we are talking about low frequency or DC. Non technical people often use the term skin effect at audio when they don't know any better.
Right?
If you don't agree, lets hear the numbers for skin effect at 100hz.
------------------------------------
If anyone thinks there is a grounding issue, they would simply connect a voltmeter between the sensor body and the frame of the car.
If you read zero volts, IT IS GROUNDED.
/** So if I hook my meter up to the chassis and coffee cup and I read zero volts, you're telling me that my coffee cup is grounded? I don't think so. Lack of a difference in potential between two objects does not gurantee that they are grounded together, though it can be an indication. Your also making the assumption that the sensor body has to be grounded and that may not be part of the requirement for the sensor in question **/

As you know, I was talking about components of a circuit. Measuring the voltage at various points in a circuit will tell you the voltage drop between those points. If those points are various ground nodes, you will see if the grounds are at the same potential. This is a fact.

The body of the sensor may not be the ground point but you must know what I am talking about. Measure the ground at the point that is supposed to be ground. If the "body" of the sensor isn't supposed to be ground measure from the point that is.
---------------------------------------
30 feet of 18AWG has .22 Ohm/30feet.
Do you really think that these sensors are driving some big load?
No, of course not. There is probably no current flow at all. It's simply a voltage level that is noted by the computer.

Even if there is 30 feet of wire with .22 ohms, it isn't going to change the voltage signal that the computer sees. The computer is only going to draw a couple of mills if that. There will be no loss at .22ohms.

And on top of that, how do you know that the sensor wasn't designed to take the connecting wires into consideration?
Sort of like zeroing out your meter leads.
--------------------------------------
My point in this whole discussion is to point out that the improvements being considered here are not needed.
No loss of performance can be shown to be caused by grounding.
Only a bunch of "what ifs" based on poor understanding of the component values we are dealing with.

The actual resistance of the steel frame is vanishingly low.

The current flow from the sensors is so low that small wire makes a good return.

RFI in sensor lines caused by ignition has no basis in fact. Measure and see.

There is no science behind the claim of extra hp.

 

[KenH]

If the ground wire at the sensor end where to bounce up by .5V due to a noise spike on the line, the output of the TPS would change by about 10% relative to the more solid ground as referenced back on the PCM.

I don't think you would get a ground bounce without a power bounce as well that is in phase.

I do think that 30' of wire is a lot of wire. And a xmas tree layout is definitely not optimal.

Since the ground wire, power and the sensor wires are not run in the same layout or run in a twisted pair type configuration, uneven noise pickup is definitely a possibility.

 

[Ulysses]

5V
|
Z
Z<----- Ground Line
Z
|
PCM Sense Input

This is the symbol representation for most of the sensors and how they are hooked up. The PCM Sensor input is indeed a low current sense line, it is of no issue. But note the ground line. What do you think?

 

[KenH]

Joe117, I don't have the time or energy to argue every single point your make. Many of your comments are valid within a certain context. Certainly some of the things that affect high frequency noise may have relatively little affect on automotive type systems. A grounding system that is perfectly adquate for grounding a DC system may prove to be inadequate when also grounding AC components on that DC system. I have never put a scope on a car ground, power or sensor lines and looked at what's running around. I would assume it is relatively low frequency with perhaps some higher frequency harmonics.

I will bet that under normal conditions, everything will look fine with a DVM. Resistance from point to point will read low and you probably won't find areas of dead metal that have a significant difference in potential between them and other grounded areas of the car with the car running. However noise pickup with this type of wiring scheme is very real. I have put put scope probes on lines and monitored noise pickup from high energy ignition sources. It is a very real event despite your claim otherwise. Wire size, length, design and distance from the noise source are all factors in the amount of coupling that goes on. A DVM is worthless for looking at that type of phenomenon.

Despite all the conjecture, I think that it has been established external to this conversation that at least some Vipers have poor sensor performance resulting in erratic operation of the car and that providing additional grounding at the sensor end of the ground wires has improved operation in this respect. That would infer that the current gounding system is a) inadequate in general or b) defective on some cars.

My question is and always has been that assuming the above statement is correct, what is the best method to improve the situation? If one were to agree that the frame is the end-all ground point in the system, then simply grounding these points to the frame makes sense. I'm just not so sure that is the case. I think this discussion needs more hard data before it can go anywhere.

 

[KenH]

5V
|
Z
Z<----- Ground Line
Z
|
PCM Sense Input

This is the symbol representation for most of the sensors and how they are hooked up. The PCM Sensor input is indeed a low current sense line, it is of no issue. But note the ground line. What do you think?

I don't have the service manual handy, but don't you have the ground and PCM sense line configuration reversed? At least that is how the TPS sensor was wired that I looked at.

 

[Ulysses]

Pretty sure,

That's what I saw last night in the schematics, but can't double check now that I am at work.

Joe,

Just out of curiosity, what would the antennae length be for signals generated from an engine RPM range say... of 500RPM to 5200RPM?

Thanks

 

[joe117]

At 5000 rpm you would have 416 spark plug firings per second.
The full wave antenna length for 416Hz would be 477 miles.

Of course, you wouldn't need a full wave antenna to couple in a 416Hz signal.

And the spark events will produce many harmonics far up into the hundreds of megahertz range.

Are any of these signals going to cause problems in the sensor lines?

I say, probably not. The sensor is sending a DC level to the computer.
The radio frequency energy would need to not only couple into the sensor wire.
It would need to be rectified to produce a change in theDC level.

The spark signal and its harmonics are very weak. A radio can pick them up, but a radio is designed to pick up weak signals.

The, somehow rectified, pulse, would need to change the dc level from the sensor. How much energy would it take to do that?
Quite a bit more than the ignition pulse can transmit.

A normal strong radio signal is about -70dBm. Converting that to voltage gives you 70 microvolts.
That's .00007 volt!!!
So, you say the signal is stronger, how strong do you thing the signal is??

Look, your cel phone doesn't make your car run poorly when you talk does it?

It is transmitting MUCH more energy than the spark plug wires could.

If someone has had a problem with the grounding of a sensor, it means nothing more than that problem needs to be fixed.
It doesn't mean that all the grounding of all the sensors is faulty and that performance will be improved by better grounds.

Guys,
take my word on this. The people who claimed that the grounding improvements could give more horsepower are simply trying to sell pseudo science to non technical people.

 

[Ulysses]

At 5000 rpm you would have 416 spark plug firings per second.
The full wave antenna length for 416Hz would be 477 miles.

Of course, you wouldn't need a full wave antenna to couple in a 416Hz signal.

And the spark events will produce many harmonics far up into the hundreds of megahertz range.

Are any of these signals going to cause problems in the sensor lines?

I say, probably not. The sensor is sending a DC level to the computer.
The radio frequency energy would need to not only couple into the sensor wire.
It would need to be rectified to produce a change in theDC level.

The spark signal and its harmonics are very weak. A radio can pick them up, but a radio is designed to pick up weak signals.

OK, thanks, I was only considering up to 3rd order harmonics. I figured it would have to be rediculously long and you confirmed it. Only thing I wasn't sure of was the strength of the radiated signal.

 

[Big Medicine]

Does this have anything to do with XM radio?

 

[Schulmann]

I had an issue with grounding.
The car was not making the expected HP and there were jumps on the dyno sheet.

Once we grounded the back Oxygen sensor and the MAP sensor we got dramatical improvements on the Dyno. The engin got "stable". I don't think that you will get more HP in the sense that you will boost the engin but you will get a stable engin with as much as HP it can make given the configuration.

 

[joe117]

Is the oxygen sensor in use by the computer during a full throttle dyno run?