Any EEC-V/Tweecer/CalEdit/CalCon knowledge here?

I am in the process of learning this stuff, and I have some questions. Thanks!

I had a TwEECer and sold it when I went to an EEC-V system. I am now using BE (Binary Editor) and a Mongoose Pro pass through cable to reflash my EEC-V box. I have been using it for over 10 years (since May 2013).

Core Tuning is the source. Gary Lewis has started using Tuner Pro and I also have another one I use but not on Fords, it is MPTune and is for Chrysler computers.

The problem I am having, at the core, is that the EEC is apparently not reading the o2 sensors. I say apparently because when logging in CalCon, no real movement of the o2 sensors is being shown.

Some info about the setup. The engine harness is a 94 Taurus SHO 3.0 MTX. The engine itself may be a 93, but I don't think that matters. It's in a mid-engine application and while the harness has not been taken apart, there are a lot of connectors or pins on connectors that are not used. No power steering, EGR, etc. The engine was rebuilt about 5-6 years ago, but has about zero miles, the car was being finished around it and it's just now ready for road trials.

In addition to the 2 narrowband EEC sensors, I have a PLX devices wideband in the right bank (what would be the rear bank in the stock transverse application). The EEC knows it as "Bank 2". Exhaust system is custom for this application so the sensors are not in the stock location, but they are about 6 inches downstream of the collector.

I have verified continuity from the 2 o2 sensor signal wires back to the EEC. The heater circuit is an add-on harness that I created, it supplies 12v whenever the key is on. The stock wiring harness was not supplying 12v to the sensors and I didn't feel like tearing into it so I just created the add-on harness.

The symptoms I am having are two-fold. First, as I stated, CalCon shows no o2 sensor activity. Second, the PLX devices wideband shows good ratios on startup, but as the engine warms up (over 30 seconds or so) the WB starts creeping towards lean at idle. Really lean, like in the 16s and 17s. Blipping the throttle brings the ratios back to the sane range. Since the EEC is not seeing signal from the o2 sensors, it does not apply correction.

So there is my first point of confusion. If the EEC doesn't see signal from the sensors, I would think the EEC would default rich and I would not be seeing the lean problem. I am not 100% positive, but I twice I thought I heard the engine sound change a bit when heading into lean. Not sure if that is my imagination, or an engine actually has a noticeable sound change when it starts to go lean, or it's a misfire.

Indeed, on one occasion when I pulled the codes after a short run, I got a "misfire on bank 2", code 419 as I recall. A misfire would certainly cause a lean condition on a WB o2 sensor. And it would explain why the EEC does not appear to be correcting the lean condition. But I have pulled the codes after a few runs, and the misfire code only showed up once, while the lean condition is constant.

I have made sure that the exhaust system is tight and should not have air leaks. I have tried fattening up the low-rpm, low-load ratios. The map I am running is just the stock one downloaded from the EEC into the tweecer, and the slightly modified one with the fatter idle ratios.

In CalEdit, I have selected HEGO1 and 2, and RPM for payloads. RPM just to make sure that there is a good datalogging connection between the tweecer and my laptop.

Here is a screen shot taken from one of the "stock" data log files that are downloaded from CalEdit/Calcon. Note that the red arrows at the bottom of the o2 vertical bars denote the low points, and the yellow arrows denote the highs, with the green arrows denoting the current position. Same for the horizontal AFR bar, you can see that there is a good swing going on.



Then here's a screenshot of my engines data logging. Nothing going on.



The green arrows dance around the bottom but don't really move. The tiny amount of movement they do exhibit is just electrical noise, I am betting.

Edit: Forgot to add that I did consider the possibility that the o2 sensors are just dead, I tried replacing them with new (TRQ OSA61297, which I believe are EEC-4 compatible, but heck I'll look at any possibility at this point). No change to the CalCon screen with the new sensors.

Any ideas welcome at this point. Thanks!

First, in any EEC system reference bank 1 is the side #1 cylinder is on, so on a Taurus bank 1 is the rear bank and cyls 123 are there, bank 2 is the front bank and cyls 456 are there. I never had a bad O2 sensor on my Taurus, but have had to replace both upper (11 and 21) on my 2011 Flex. FWIW, these are a dual band sensor of some sort.

If by chance you have the sensors crossed I know you will get some very strange results.

I have run into some issues with Bosch O2 sensors not working on Chrysler products, I don't know if maybe you are seeing something like that.

Good luck with it

85lebaront2 wrote

First, in any EEC system reference bank 1 is the side #1 cylinder is on, so on a Taurus bank 1 is the rear bank and cyls 123 are there, bank 2 is the front bank and cyls 456 are there.

You are correct, I misidentified the banks. So the WB o2 is in bank 1, not 2. So apparently the misfire event occurred in the bank without the WB o2.

85lebaront2 wrote

I never had a bad O2 sensor on my Taurus, but have had to replace both upper (11 and 21) on my 2011 Flex. FWIW, these are a dual band sensor of some sort.

If by chance you have the sensors crossed I know you will get some very strange results.

The sensors on your Flex were dual band of some sort, you are saying? I'm pretty sure WB 02 was a thing by 2011, but I'm relatively certain that anything from the 80's to mid-90's was a narrowband. I'm familiar with NB and WB sensors, but I'm not sure what a "dual band" sensor is.

I'm not sure what you mean by "have the sensors crossed". Are you referring to the possibility that I have the wrong sensors?

85lebaront2 wrote

I have run into some issues with Bosch O2 sensors not working on Chrysler products, I don't know if maybe you are seeing something like that.

I am on my 2nd set of sensors from different brands. The first set -may- have been dead but I doubt it. The second set (the TRQ set I referenced) I bought from PartsGeek and their website definitely shows that product as compatible with 89-95 Taurus SHO. I guess I will throw in the Denso 234-3002's I have (brand new) and see if it makes a difference but I am am giving that about a .01% chance of making a difference.

Thanks for the reply.

Bank 1 by SAE standards is the bank with #1 cylinder, bank #2 is the other bank. You correctly identified them in this post. The Flex sensors have 5 pins but 6 wires and the NTK catalog I have shows them as having two sensors in them. I used the Motorcraft as they were actually reasonably priced. The real fun, in 2011 there were 3 different upstream O2 sensors depending on when it was built.

Basically you need to be sure that bank 1 has O2 sensor 11 and bank 2 has O2 sensor 21 Those are the standard numbering system for more than 2. with only 2 they may be designated as 1 and 2.

The normal O2 sensor is a narrow band and is only effective in the 14.6 to 14.8:1 A/F ratio, or a Lambda value of 0.99 to 1.01 if your mixture as measured by your wideband is lower or higher, the sensor will just be stuck at it's range limit which seems to be what you are seeing.

Here is something I found that may help:
What is the difference between a wideband O2 sensor and a narrowband O2 sensor?
Narrowbands are excellent when it comes to finding 14.7:1, but they aren't nearly as accurate when air/fuel reaches either side of this ratio. Wideband sensors are built for pinpoint accuracy. They send voltage to the ECU on a scale of 1 to 5 so that it knows exactly what is required to maintain target air/fuel ratios.

85lebaront2 wrote

Bank 1 by SAE standards is the bank with #1 cylinder, bank #2 is the other bank. You correctly identified them in this post. The Flex sensors have 5 pins but 6 wires and the NTK catalog I have shows them as having two sensors in them. I used the Motorcraft as they were actually reasonably priced. The real fun, in 2011 there were 3 different upstream O2 sensors depending on when it was built.

Basically you need to be sure that bank 1 has O2 sensor 11 and bank 2 has O2 sensor 21 Those are the standard numbering system for more than 2. with only 2 they may be designated as 1 and 2.

OK so I guess that's what you were talking about when you were referring to crossing the sensors - you meant using a sensor designed for forward operation in the rearward position, or vv, is that correct? This engine has only 1 sensor per bank, so that's not it.

85lebaront2 wrote

The normal O2 sensor is a narrow band and is only effective in the 14.6 to 14.8:1 A/F ratio, or a Lambda value of 0.99 to 1.01 if your mixture as measured by your wideband is lower or higher, the sensor will just be stuck at it's range limit which seems to be what you are seeing.

I understand that a NB sensor can only function right around stoich and that they function by switching on and off, they are mostly a binary thing, either sending about 1v or not sending 1v. From the web:

"A narrowband sensor works on a rich/lean principle. When the air/fuel mixture is on the rich side, the voltage output is around 0.8 to 0.9 volts, and when the air/fuel mixture is lean, voltage drops to 0.3 volts or less."

Although this seems counter-intuitive, the way a NB works is that it has an inner bulb exposed to atmospheric o2, and the outer bulb is in the exhaust stream. The voltage is generated from the o2 difference, not the o2 in the exhaust.

So the voltage is being sent in the absence of oxygen in the exhaust. The sensor can see 3 different conditions:

1. Stoich - air and fuel are consumed in equal parts, leaving no oxygen (sensor shows voltage - no oxygen present)

2. Rich - the combustion event consumed all the o2, but there was fuel left over, so like #1, no oxygen left (sensor shows voltage - no oxygen present)

3. Lean - the combustion event had oxygen left, so the mixture did not have enough fuel to achieve stoich. Since there was oxygen left over, there was no o2 difference detected between the inner and outer bulb of the sensor, thus no voltage.

Having said all that (and a large part of that was just me thinking through it), then I could see that if the engine remained at idle, if the lean condition was real, there would be no voltage coming from the sensor and it would be "stuck" on the lean side. However, I don't let the engine idle in the lean area, I rev it up. And the WB sensor shows 12's and 13's then which should be generating voltage, and should make the readings jump up. I realize that the sensor doesn't start to work until it's about 600F, but I have gotten the coolant temp up to 180 and still no readings.

One experiment I have thought of is to take a small 1.5v battery and resistor it down to about 0.9v. Then feed that into the EEC and see if CalCon shows activity on the HEGO bars then. That might isolate whether the sensors just aren't putting out, or if there is an actual problem with the EEC or the CalEdit/Calcon datalogging payload or software.

Pete Whitstone wrote

One experiment I have thought of is to take a small 1.5v battery and resistor it down to about 0.9v. Then feed that into the EEC and see if CalCon shows activity on the HEGO bars then.

I'm sure you are aware, but just in case: There has to be current flow in order to have a voltage drop across a resistor.  The equation is E (the voltage drop) = I (the current) x R (the resistance).

So if you are wanting .9v from a 1.5v battery you'll need two resistors in series with the bottom one twice the value of the other.  Said another way, if you want the current to be .15A then R = E/I = 1.5/.15 = 10 ohms, total.  So you'd want a 3.3 ohm resistor connected to the positive of the battery and a 6.6 ohm connected from that one to the negative of the battery.  And at the junction of the two resistors you'll have .9V.

But 150 ma might be too much for a AA battery, so you could go with 100 ohms total and the current would be 15 ma.  But you need the current through the resistors to be at least an order of magnitude larger than whatever your EEC pulls to ensure it won't alter the voltage.

Thanks Gary, that's mostly Greek to me. I have a friend who's an EE so I was going to seek his help. Can you draw a quick schematic of what you are talking about? I think I could understand that better. I can interpret the word version a couple different ways in my mind. Thanks!

Not pretty, but here's what I was thinking.  R1 is always 2X R2, but the values can vary to decrease the current.  Again, the current needs to be at least an order of magnitude more than what the ECU pulls, although that is probably very, very little.

Thanks for the visual, Gary. I thought that was what you were saying but the graphic confirms it.

I'm curious about the part where the current (amps?) needs to be 10x what the ECU wants. Can you explain that a bit more?

The current is what sets the voltage. So if the current pulled by the ECU is an appreciable portion of that pulled by the resistors the voltage will change as it only pulls through R2.

Ah, gotcha. So if the current available from the battery/resistor stuff is only a bit more than the EEC draws, it will drag the voltage down to next to nothing. But if it's 10x more than the ECU needs, it will only drag it down about 10%.

Makes sense. Thanks!

Sidebar question - would I need 1 3.3 ohm resistor and 1 6.6 ohm resistor? Or would 3 3.3 ohm resistors work, one on the + side and 2 in series on the - side?

You can use three resistors of the same value, which is a good idea as it keeps it simple.  I have ordered from Digikey for many years and have this search set for the kind of resistor I think you want, through-hole, and 1/4 watt.  And there I see 10 ohm resistors as an option.

So let's do the math.  3 x 10 = 30 ohms total resistance.  I (current) = E (voltage)/R (resistance) = 1.5/30 = .050 amps, or 50 milliamps.  Wattage, meaning the power that the resistors will dissipate, is I x I X R = .05 x .05 x 30 = .075 watt.  So you'll be fine with a resistor rated at .250 watts.  In fact you could go smaller if you wanted, but 1/4 watt resistors are probably the cheapest as they are quite common.