Wideband sensor plug too large

[eflight]

Got a wideband sensor and an AFR gauge for my 84 Targa
In the process of installing the sensor wire and found that the plug is too large for the hole that's in the engine tin.

Anybody else run into this and have a good solution or ideas before I start cutting the engine tin.

The sensor is about 1" dia so that may be the easier way to go, just need enlarge the hole if I go that way

[HaroldMHedge]

I would try to de-pin the connector, run the wire thru and put the pins back in the same locations.

[eflight]

Originally Posted by HaroldMHedge

I would try to de-pin the connector, run the wire thru and put the pins back in the same locations.

thought about that, but could not get the connector apart without breaking it

[76FJ55]

You'll probably be a lot close passing the sensor through the hole as opposed to the connector. It still won't quite fit but is pretty close. On my previous 928 I filed the round hoe into a hex that matched that on the sensor. The factory grommet had a large enough lip that it still fully covers the hex shaped hole.

[Mahler9th]

I think it is best to make the hole in the tin larger.

If you do that, you may be able to buy or make a grommet for that hole if you are concerned about heat, debris or whatever.

[spuggy]

Originally Posted by eflight

thought about that, but could not get the connector apart without breaking it

Most modern terminals lock into the connector body like this:




There are many tools to de-pin connectors. Some generic, some for specific connector/terminals. A trivial task with the correct tool.

This is the connector kit for a Bosch 4.9 wideband (the 4.2 wideband has the same terminals - but the connector body is deliberately different/incompatible, so you can't mix them up - as they require different controller circuits).

Hard to tell what exactly the terminals are, but given the Bosch origin, I'd lean towards thinking they're some variation of TE Superseal - there's at least 3. The terminals look similar to GM/Delphi Weatherpack, but are supposedly different (Weatherpack/Superseal connector bodies don't look similar at all).



These are female terminals, (and facing in the wrong orientation in the picture, the weather seals are supposed to be crimped into the wide part of the connector to seal the hole in the connector body) - Note the tang, or barb, on the right side of the terminal.

You release the barb from the front with the tool, and then pull the wire back through the back of the connector housing. It may help to push the wire from the back to reduce any loading on the barb/make sure it's got room to move out of the way,

"Assembly is the reverse of disassembly", as all the best manuals say, LOL.

You can often feel a positive "click" as the barb locates into the body. Although the wire not pushing back through the connector body is also good enough.

[A930Rocket]

Loosen the tin (or lower the engine) and run the wire between the tin and the engine seal. No need to modify the tin. That’s what I did with my 930.

[Betis914]

Just make sure to clean up the edges and maybe add a grommet to keep the wire from chafing.

[TopTuning]

The sensor fits through the hole. The connector does not.

[Mahler9th]

Many or uego devices use sensors that have a built in trim resistor. If this is the case, instructions usually caution folks about altering the harness to which the sensor connector is attached. If you cut and splice in wires to extend the length of this harness, you may compromise the accuracy of the readings.

So if you feed the sensor through the tin, and there is not enough harness length to get the sensor into the bung, it may not be a good idea to lengthen that harness.

[spuggy]

Originally Posted by Mahler9th

Many or uego devices use sensors that have a built in trim resistor.

A wideband UEGO sensor is a narrowband sensor with two other things added; the heater circuit used to maintain the correct temperature range, and an electron ion oxygen pump - to provide reference air to compare the exhaust gas sample against.

All the Bosch wideband sensors incorporate an internal resistor pack laser-trimmed at the factory. If you buy a Bosch wideband from MoTeC, they helpfully etch the initial value on the body of the sensor.

A MoTeC allows you to perform a free air calibration to measure the current value of a not-new sensor. Which you need to do periodically to compensate for the value of the sensor varying as it ages and/or accumulates contaminates. Any system that doesn't allow you to re-calibrate the sensor doesn't seem like it would be fit for purpose to me.

You can actually measure this value with a DMM between a couple of pins on the connector (1 & 6 on my 4.9), multiply by 10 and arrive at the same value as the value the MoTeC reports from the free air calibration step. Does for me, anyway.

Originally Posted by Mahler9th

If this is the case, instructions usually caution folks about altering the harness to which the sensor connector is attached. If you cut and splice in wires to extend the length of this harness, you may compromise the accuracy of the readings.

It's not the resistance of the wire that's the problem.

What Bosch are actually concerned about is that owners don't solder the crimp and/or heat shrink the wire harness. The Nernst cell doesn't/can't work without a source of reference air - and in the Bosch design, no port is provided on the sensor body itself. Because they judged that it would be too vulnerable to being clogged with road crud. So, in the Bosch wideband, the ion pump pulls the reference air it needs down the strands of the wire in the harness.

Many folks have expressed skepticism about this fact, despite it being called out in almost every Bosch tech document - for example, Bosch LSU 4.2 Tech Doc Y 258 K01 005-000e Mar 2001 contains this text:

9.13 For physical reasons the sensor needs ambient air at its reference gas side. Replacement of the air volume inside the sensor must be guaranteed by a sufficient air permeability of the wires and the connectors between sensor and ECU. The breathability should be higher than 1 ml/minute at a test pressure of 100mbar.
The current Bosch LSU connector is available with an integrated pressure compensation hole with a permeable membrane, which guarantees a sufficient air supply. This hole must be saved from deterioration as wax, oil etc.

and

9.15 The PTFE formed hose is part of the reference air volume of the sensor and must be kept sealed and undamaged. For installation, the minimum bending radius of the hose must be 20mm (for long PTFE hose) resp. 12mm (for short hose). Keep the PTFE formed hose away from sharp edges and avoid contact/friction with frame/engine assembly.
The first fixing point for the cable to the car body should be 200mm to 400mm after the end of the PTFE formed hose, depending on movement of the exhaust system.

Bosch sell "universal" sensor kits, with unterminated wire ends, and provide a solderless wire nut type connector (special high-temperature rated Posi-LockŪ) with this - and instructions that explicitly spell out not to heat shrink or solder the wires - or to extend the wire harness sheath.


NTK also make wideband sensors - but I've no idea how they arrange for the reference air flow, as I only buy Bosch...

[scarceller]

Where are you installing the actual gauge? Somewhere in the cockpit? If so you do not need to go up into the engine bay. If it's a 84-89 Carrera you should route wiring along side the main engine harness that enters the cabin below the rear driver's side jump seat.

What WB02 Gauge did you purchase? More details please.

[eflight]

Thanks all for the inputs

Sensor is the AEM 30-4110

No the sensor does not fit through the hole as it is.

I broke out the trusty step dill and opened it up just enough to slide the sensor through the larger hole.

A little RTV using the old grommet from the old O2 sensor and its done.

I did run the AFR wire out the front of the engine bay and down into the grommet in the floor pan behind the drivers seat. Also ran a spare wire in case I wanted to use the O2 sensor emulation from the gauge.

Getting that grommet back in place afterwards was all kinds of fun.

I ran a 12V line from the fuse panel.

It all got put in a mount on the center console that I printed out. New panel has a voltmeter, USB socket and switch for the antenna so I can have it stay down if just listening to the music on the thumb drive.

The good news is that it all worked.

The bad news is that it worked for all of about 20 seconds. I took a picture of it after I backed out of the garage. As I pulled out of the driveway it went to 3 dashed red lines and stayed that way. Turned off the car, restarted it, goes through its calibration sequence and then 3 dashed lines.

Called AEM, they are sending me a test procedure to figure out whats wrong.

[Mahler9th]

As I have posted in other threads, it can be important to be careful about oxygen sensors for AEM and other products.

A bunch of us used AEMs for many years in racing apps using 110 leaded. The sensors lasted a long time for quite a few years... but then things changed.

And I bought an AEM "compatible" sensor or two from Amazon and/or Ebay back in the day with poor results.

I still have several AEM sensors that are like new if you want them for trouble shooting.

Our "herd" switched to the Ballenger AFR 500.

We always compared to an old Motec "gold standard" from back in the high days of IMSA.

Again application = racing and running open loop and leaded fuel.

Great company and products.

I have been to the NGK/NTK silly con valley office in SJ. Great people preparing for a different future.

[eflight]

Talked to the rep at AEM today. They have 3 tests to do before they can determine what's wrong with AFR system.

Did test 1. Run positive and neg wires directly from gage input wires to battery.
I did that and the AFR System starts up and within 10 seconds is showing full lean. thats with the engine off.

Test 2 is to see if sensor heats up, which means jacking up the car, taking off the wheel and pulling the sensor.

Test 3 is gas rag for a sniff test on the sensor after test 2.

Looks like lots of rain this week so it will get done when I cant do anything else