I have been thinking it might be a good idea to get the various grounds tied in together for the EFI. This brought up several questions.

1. Is the O2 sensor just a variable resistor to ground? One wire systems have no separate ground wire. Some wiring diagrams of the MS ECU show the second wire being taken to MAT "return" tap on the ECU. There are a number of these "return" taps on the relay box. If the O2 sensor is just a variable ground source does it matter which wire goes to ground and which one goes to the relay box tap,(when there are two wires)? Mine also has two white wires which are the heater B+ and ground.

2. Are these sensor return taps in the MS relay box just grounds? It seems they must be. There are at least three of these,and maybe more. If the box is grounded to the same place as the sensors,(which they are not now) how does this effect the signal and performance of things?

3. If the signal return taps are grounds, is there any advantage to going to the box versus just tapping into the same ground spot on the bulkhead?


4. Right now my ECU is grounded to the inner fender, and the sensors are all grounded to the engine block, (well, some of them).
Thanks guys.

Sam

_________________

Sam,

You can do a lot of reading about ground loops. You've created ground loops by not having the sensor and ECU grounded to the same place. Ground loops can cause problems. http://en.wikipedia.org/wiki/Ground_loo ... tricity%29

1. The O2 sensor is a voltage source.

2 & 3. Yes, the sensor returns are grounds, but the ground needs to be at the exact same potential as the ECU for the sensor output to be accurate at the ECU. If the grounds are not at the same place the ground potential can differ.

4. Not optimum. Bring all the grounds together at one point.

_________________
Joshua

I would also recommend making that point either on the engine block, cylinder head, or the battery negative terminal. Grounding the ECU to the sheet metal can cause noise issues.

_________________
"Mad Scientist" Matt Cramer
'66 Dart - turbocharged 225
My blog - Mad Scientist Matt's Lair

Thanks for taking the time to reply guys. Especially in the middle of the day like that.

Can the O2 sensor supply a voltage without an outside source? Doesn't the voltage have to come from somewhere else? For instance, the TPS has a reference voltage of 5 supplied to it from the ECU and the ECU then reads a return voltage that results from the variable resistance of the TPS position.

According to the O2 sensor wiring diagram there is no outside source of power, unless the O2 feed itself puts voltage onto the sensor. The way my O2 sensor is currently wired the black wire is to the O2 signal tap on the relay box terminal strip, which is number 20, and the last one to the right . And the grey wire is grounded to the engine head. The remaining two white wires are split between ground and B+. This works. The lap top tuning program shows the O2 sensor feedback as working correctly when wired this way.

It seems the O2 sensor itself is simply a resistor in series with the O2 circuit, and the resulting signal is simply a reading of the voltage this circuit will carry at any given moment. Granted my understanding of such things is shaky at best. I'm just throwing this out there to test my understanding.

To add to this, one of the diagrams I have seen on DYIautotune's web site shows the grey wire going to the CLT ret tap on the fuse block. This suggests that the taps marked ret for CLT, MAT and TPS are all grounds. So, if these taps marked RET are grounds, is this not the best place to bring the sensor grounds back to the ECU? If you ground the ECU to the head, through the relay box, and then ground the sensors to the same tap, is this in effect the same as bringing the wires all the way back to the relay box? If the answer is yes, this eliminates some wires under the hood, which there are plenty of now.

As I have said many, many times, I am not an EE by any stretch. So, don't hesitate to correct me here. Just be polite about it.

Sam

_________________

The O2 sensor generates voltage. It's unlike any other sensor used in the engine management system. http://en.wikipedia.org/wiki/Oxygen_sensor The two white wires are only for the heater circuit. The sensor will still work with the heater disconnected, but only after the exhaust gas has heated the sensor sufficiently. The gray ground wire is more reliable than a sensor grounded to the manifold through the shell.

Yes, all the sensor returns are grounds and they should all terminate at the same place. You can bring several grounds together pretty much wherever you like and then take them to their final ground point. Just avoid loops in the ground circuit. If you draw a schematic of your ground system and find multiple paths to from any component to ground then you've created ground loops.

_________________
Joshua

Thanks Joshua. I appreciate your help. That is interesting indeed. Ingenious! So this means if I want, I can read that voltage right off the wire with a VOM I guess. You might consider me a special needs EFI guy.

(You were indeed polite. Thanks)

Sam

_________________

Only if you touch the black lead from the VOM to ground..............

_________________
Ed
64 Valiant 225 / 904 / 42:1 manual steering / 9" drum brakes

Sam,

You're welcome. Yes, people have used a volt meter to monitor their O2 sensor. There was a turbo Dodge guy that used to sell what was essentially a little volt meter with LEDs which were supposed to correspond to the A/F ratio, but those meters aren't accurate because the A/F to voltage curve is not linear except right near Lambda. With wideband O2 sensor and gauge packages as cheap as they are I can't see trying to tune with a narrow-band.

_________________
Joshua

It's just something to do until my wide band controller comes back from Inovate.

Sam

_________________

This is the most accurate voltage vs AFM for a narrowband I've found:


http://teamzr1.com/o2volt.htm

_________________
Ed
64 Valiant 225 / 904 / 42:1 manual steering / 9" drum brakes

Ed,

That's great info. Thanks for posting the link!

_________________
Joshua

Thanks Ed. It appears that the AF ratio can vary by as much as 2 points depending upon the EGT. I guess lacking hard data, it would be safe to figure on the high side right behind a turbo. In fact, maybe even higher. Matt Cramer says the EGT must be 900 degrees or lower for a wide band to survive. And they suggest at least 18" down stream from the turbo for that.

Oh, I just noticed, that is centigrade and kelvin in the chart. I will have to do some conversion. Thanks for the info. I will now have something to do while I wait for Inovate.


You can tell from looking at those numbers why narrow band sensors are not a very reliable read of AF ratio. And, the hotter the exhaust, the more useful the numbers. At the hotter end of the range variance from 14.7 shows up as voltages that are closer to each other. When the exhaust is cooler just off stoich is .942 on the rich side, and .101 on the lean side. This compresses the values considerably for the remaining AF ratios both lean and rich. However with a very hot exhaust the lean side of stoich is .153 and the rich side is .659 meaning the remaining voltage values from there to 0 and 1 are spread out more.


Sam

_________________

That O2 stuff is from Heywood's book - a bible of engine physics/engineering from MIT professor.

Lou

_________________
Home of Slant6-powered fun machines since 1988