Notice: We request that you don't just set up a new account at this time if you are a previous user.
If you used to be one of our moderators, please feel free to reach out to Chris via the facebook Outerlimits4x4 group and he will get you set back up with access should he need you.
Recovery:If you cannot access your old email address and don't remember your password, please click here to log a change of email address so you can do a password reset.

Oxygen sensor operation in Ferozas....

Tech Talk for Ford, Mazda, Daihatsu & Makes that currently dont have a home.

Moderator: Tiny

Post Reply
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Oxygen sensor operation in Ferozas....

Post by murcod »

The first part in my spiel on oxygen sensor operation....

The oxygen sensor is located in the exhaust system between the exhaust manifold and the catalytic convertor. It is used to control the air:fuel ratio in the HD-E EFI engine to a value of 14.7:1. This ratio is chosen because it is where the engine operates most efficiently (power vs fuel usage) and where the catalytic convertor works best at converting harmful exhaust gases to less toxic emissions.

Basically the oxy sensor operates by generating a voltage relative to the oxygen content in the exhaust gases. This output voltage is directly related to the air fuel ratio being fed into the engine. To operate correctly an oxygen sensor must get up to it's operating temperature (around 400 degrees Celsius) before it starts generating a meaningful output voltage.

Some oxygen sensors employ a 12v heater to get them up to operating temp quickly to help meet strict emission standards. The Feroza, having an basic/older EFI system, employs the simplest form of oxy sensor- a single wire non heated narrow band type. Narrow band refers to the fact it is only capable of giving an accurate output around the 14.7:1 air: fuel mixture range (ie you cannot use it's output to accurately determine what the exact A:F ratio is outside that- only that the engine is running rich or lean). The single electrical wire is used to feed the sensor output voltage to the EFI computer and the oxy sensor's stainless steel body relies on the exhaust system being grounded to the engine (which is in turn grounded to the battery negative) to complete the electrical circuit.

Not being heated, it relies on the exhaust gases to get it up to operating temp. Observing the output voltage on a multimeter from a cold start reveals the voltage starting off at around 0mV and rising slowly as the engine warms up. The operating voltage range of the sensor is from 0mV (0.000V) up to around 900mV (0.900V) DC. Once the oxy sensor is up to operating temp a reading of around 0mV means an extremely lean air fuel ratio mixture and 900mV indicates a very rich mixture. The voltage output for the ideal 14.7:1 air: fuel ratio is around 450mV.

In practice, once the oxy sensor is up to temp and the EFI computer is using it's output to control the air fuel ratio, the output voltage will be observed to be oscillating up and down many times per second around the 450mV ideal. This shows the EFI computer is operating in "closed loop mode" -meaning it's using the oxy sensor output voltage to control the air fuel ratio to 14.7:1. The mixture is changed by the EFI computer varying the time the fuel injectors stay open, thus supplying more or less fuel to mix with the incoming air.

Under full wide open throttle conditions, or when the throttle is closed fully (decelerating), or when the oxy sensor isn't up to operating temp the EFI computer uses data stored within it's memory to control the mixtures and ignores the oxy sensor's output. When operating in such a manner the EFI computer is said to be operating in "open loop mode."

Next will be some of the symptoms that a broken oxy sensor can cause....
David
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

Here's some things off the net that a dead/ dying oxy sensor can cause:

Surging and/or hesitation
Decline in fuel economy
Unacceptable exhaust emissions
Premature failure of the catalytic converter

From the info I found an oxy sensor can still be outputing the correct voltages but not be serviceable and cause problems. As they age the output voltage response time to mixture changes can increase resulting in the EFI computer changing the injector pulse width out of sync with what is really happening. This will cause the hesitation/ surging as the power output will vary with the air fuel ratio mixtures going all over the place. This was the problem with mine (after I'd fixed the earthing problem!)

Another point with the Feroza is the EFI fault light will not light on the dash if the sensor output goes open circuit! So it's quite feasible to be driving around oblivious to the fact- like I was for over 12mths. Power output will be down and fuel consumption will be higher. In my case the exhaust system was somehow not electrically connected to the engine- corrosion must have built up between the bolts and exhaust manifold / head.

To be continued....
David
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

Things that can kill an oxy sensor:

- using silicone on the engine eg. to seal gaskets; only use "Oxy sensor safe" silicone
- using leaded fuel
- dropping them!
- engine lubricants, sealants and additives, and/or oil can damage the sensor's element
- phosphorus from excessive oil being burned in the combustion chamber
- coolant leaks (eg head gasket) as some coolants contain silicone
- age; they are only supposed to last 80000- 130000km
- being externally coated in mud oil etc; the rear of the sensor must be vented to the atmosphere in order for it to operate
David
Posts: 2431
Joined: Thu Sep 04, 2003 12:02 pm
Location: Stayton, Oregon, USA

Post by SimplyPV »

wow.... i seriously need to replace mine then... 120,000 miles.... TONS of mud.... oh and that would explain the engine light coming on and off.... great job david! i'm impressed..... now a question for you, with my mainifold being cracked like hell, should i replace the oxy sensor or should i wait until i get my new extractors fitted? which will be a while anyways due to $$ issues....
[quote="simplypv"]its a Strine thing and i just dont understand![/quote]
Regards, PV
Team [size=134][color=red][i]Anti-[/i][/color][/size]Asshat [b][color=orange]#5[/color][/b]- Yank!
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

PV I'm going to get around to writing a bit more on how to fault find the oxy sensor. Need to take some pics though first.

It's relatively easy to check them with a multimeter.
David
Posts: 4323
Joined: Wed Dec 11, 2002 8:42 am
Location: Gold Coast, Queensland

Post by HeathGQ »

what a wealth of knowledge you are David. you amaze even me sometimes. Well done. :armsup:
Heath & Melissa - 93 GQ LWB.
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

I was going to wait until pic posting was working again; but I'll make a start now.

Some basic oxygen sensor checks ccan be carried out with a digital multimeter (DMM). The DMM doesn't have to be a flash one and these days you can pick them up for under $20 from Dick Smith etc and get a reasonable one.

Starting under the bonnet, first locate the oxygen sensor in the exhaust manifold. It's the metal thing sticking out with the single wire coming out of it.

Image

Using the multimeter set on the lowest resistance range connect one probe to the negative battery terminal and the other to the stainless steel body of the oxygen sensor. You should get a low reading of no more than a couple of Ohms (ideally a reading of zero.) This is important as only having a single wire on it, the oxy sensor relies on the earth through the exhaust manifold-> engine -> chassis -> battery negative (earth) for it to function. Without this earth completing the electrical circuit the EFI computer will not be able to read the voltage the oxy sensor is producing.

If you don't get around zero Ohms try to isolate where the problem is by checking the manifold with the probe to see if it is earthed, then the engine itself etc. Don't be afraid to "dig" around with the probe to get through rust to bare metal- rust won't conduct electricity and may give you a false high resistance reading. My Feroza had an oxy sensor earthing problem- so don't just assume yours is OK!
David
Posts: 576
Joined: Wed Jun 02, 2004 8:17 pm
Location: Adelaide, South Aus

Post by Dangerous Dave »

David,

If i haven't said it before, i'll say it now, YOU ARE A CHAMPION. If only all workshop manuals etc were as easy to understand as this thread. Funnily enough i put extractors on my car a little while ago and used silastic on the exhaust manifold gasket. A little time after that the Roza started to run like thpppp. Makes plenty of sense to me. I picked up a new o2 sensor yesterday. The NGK one, i got it for $63. Will put it in today along with a new set of plugs and see how things go.

Cheers

Dave :armsup:
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

Good to see it's already helped someone. :D

I think a few people will also find their sensor isn't earthed if they've got the original manifold. I'll try to add a bit more on it later today.
David
Posts: 2431
Joined: Thu Sep 04, 2003 12:02 pm
Location: Stayton, Oregon, USA

Post by SimplyPV »

ok now i'm confused... if it wasnt earthed before.... on the stock mainifold that is... then why has it worked well up until after it starts conkin out???
[quote="simplypv"]its a Strine thing and i just dont understand![/quote]
Regards, PV
Team [size=134][color=red][i]Anti-[/i][/color][/size]Asshat [b][color=orange]#5[/color][/b]- Yank!
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

Could have been the same as mine- not earthed on the stock manifold. Fit extractors, now it is earthed and working and you start to notice running problems because the mixtures are all over the place. The silastic wouldn't have helped if it wasn't oxy sensor friendly stuff.
David
Posts: 2431
Joined: Thu Sep 04, 2003 12:02 pm
Location: Stayton, Oregon, USA

Post by SimplyPV »

hmmm... well i know mines getting worse.. its starting to really choke up from time to time when i give it gas at a stop.... and gas mileage is seriously going to shit now.... and i'm gettin worried... sadly i cant do nothin until AFTER my exhaust is replaced.....
[quote="simplypv"]its a Strine thing and i just dont understand![/quote]
Regards, PV
Team [size=134][color=red][i]Anti-[/i][/color][/size]Asshat [b][color=orange]#5[/color][/b]- Yank!
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

Have you done the earth check PV?
David
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

There are a couple of ways to check the Oxy Sensor's output voltage, but they all involve measuring the output voltage and seeing how the output is responding. This means having to hook up a multimeter somewhere into the sensor output wire while the engine is running.

I found the easiest place to monitor the voltage was inside the cabin at the EFI computer (also call the ECU- Engine Control Unit) connector. The ECU is located in the passenger side footwell, below the glovebox (talking RHD Feroza's here BTW ;) ). It's the aluminium box that's bolted onto the side panel with the two large electrical connectors going into it. There should be a protective black plastic cover fitted over the connectors- remove the 10mm bolt that holds it on and you'll be able to see something similar to the photo below. (Just ignore the red and green wires hanging down in the photo below... you won't have them on yours yet.)

Image

The easiest way to hook into the wiring to monitor voltages is to insert a length of wire into the back of the connector shell. Solid core wire (about 1mm conductor diameter) is best for this; or you can use multistrand wire and solder the ends like I did. You'll need two lengths of wire about 30cm each with about 1cm of insulation stripped off each end.

Below is the ECU pin out diagram:

Image

As you can see we're interested in pin 29 (Oxygen Sensor) and pin 21 (Engine Ground). The next step is to insert the two wires into the connector- this is easiest to do with the ECU connectors unplugged. Do this (with the ignition OFF) by gently squeezing down on the raised release tab on each connector and then pull the plug out.

Now grab a pair of needle nose pliers and insert the stripped wire into the back of the plug as show below. The Oxy Sensor wire is white in colour, and the Engine Earth wire is brown with a blue "ring" on it. The test wires should slide in with a bit of wiggling and sit wedged in the connector shell next to the electrical pin.

Image

Image

It is important to make sure that there are no wire strands from the test wires shorting out on the pin next to the one being monitored. This could happen if multistrand wire is used without the ends soldered.

After the wires are inserted, insulate the other end with tape so it can't short out and plug back in the two ECU connectors. You should end up with something like this:

Image

It's important to not allow the Oxygen Sensor output wire to short onto either the vehicle body or the Engine Ground wire. Damage to the Oxy Sensor could result (-unlikely, but it pays to be careful.)

The next step is to hook up a digital multimeter (DMM) onto the test wires. Small alligator clip leads are handy to do this. Once again wrap all joins in insulation tape to prevent shorting. The multimeter should be set on Voltage and the range to read 0->1V.

Image

With engine off there should be close to nothing, around 0 millivolts (mV), or 0.000 V output on the DMM.

You can now start the engine. Initially the voltage reading will be very low and then it will slowly increase as the exhaust temperature (and Oxygen Sensor) warm up. Remember Oxy Sensors need to be around 400 degrees Celsius to give a useful output.

The best way to test is to get somebody else to drive while you watch the multimeter. After a couple of kms driving the Oxygen Sensor output should have increased to around 450mV (0.45 v). At this stage try to find a section of road where you can sit on a relatively constant speed - around 60km/h in fourth gear is good. The driver should avoid lifting off the accelerator completely or flooring it. You should then be able to see the multimeter reading constantly jumping up and down around the 450mV region. This indicates the ECU has gone into Closed Loop operation and is using the Oxy Sensor to control the mixtures. It will never sit exactly on 450mV as the mixtures are constantly changing; but the average voltage should be approx 450mV.

You will notice if the driver lifts off the accelerator completely the voltage should drop straight away to a very low reading- indicating the injectors have been shut off and a very lean mixture. If the accelerator is floored the opposite should immediately happen with the voltage jumping up to around 900mV or more (indicating a very rich mixture.) It is important that the voltage reacts immediately to either situation- even half a second delay indicates a dead Oxygen sensor and will cause driveability problems (surging, hestitation etc.)

If you think there is a delay problem then stop driving and open the bonnet. With the engine revs held up around 3000RPM disconnect the fuel pressure regulator vacuum hose. See the below pic for the location of this hose.

Image

This has the effect of instantly increasing the fuel pressure (giving a rich mixture condition) and should immediately give over 900mV on the DMM. If there is a delay then the Oxy Sensor should be replaced.
David
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

Final part.

Where can you get a new Oxygen Sensor???

Many companies make aftermarket oxy sensors to suit the Feroza eg. Bosch, Denso, NGK (under their NTK brand) etc. Beware though that anything labelled "universal fit" will not have the correct waterproof connector plug on it- only a bare spade type crimp terminal. This means you run the risk of dirt and water causing problems with a high resistance electrical connection and more headaches down the track- so pay the extra and buy one with the proper waterproof plug! ;)

I found NGK ( www.ngk.com.au ) only listed a universal one to fit Feroza's- however they list one with the proper connector to fit the Applause HD-E, along with the same universal fit one. I ordered the Applause one and discovered it will fit the Feroza loom- the plug is the correct one. The only difference is the wiring between the sensor and the plug is around 50cm long, compared to around 20cm for the standard Feroza sensor. This is a bonus if you've extractors fitted because the factory sensor's lead is way too short. :)

The part number for the correct NGK sensor is : OZA401-E70 , this is the Applause fitting with the correct waterproof connector/ plug. RRP is around $85 AUS; but you should be able to pick one up for a bit less.

Note: When you buy the above sensor it will come fitted with a bracket (shown below)- this is not needed for the Feroza. Carefully place the bracket in a vice, as shown, and use a spanner to unscrew the sensor. It will then bolt straight into the Feroza.

This is my old oxy sensor screwed into the bracket that comes fitted on the NGK OZA401-E70

Image

Here's what the sensor looks like after removal from the bracket

Image
David
Posts: 2
Joined: Sun Jun 27, 2004 10:57 am

Post by aus.leo »

Do you have a photo of the actual plug (connection at the end of the wire) for the o2 Sensor, showing which wires are connected to it. Mine has been disconnected, I am unsure which is the correct wire to connect to the o2 Sensor ??? and have found some cut wiring near the firewall.

edited; can i just hardwire from pin 29 direct to the o2 Sensor by splicing the wire at each end (ECU plug + o2 wire)
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

Welcome Aus.leo, good to see you've joined Outerlimits. :)

The plug it connects to normally sits in some clips attached to the firewall end of the head/ block (just below the distributor.) The clips you're looking for are similar to the ones the distributor's connector clips into.

If you look into the connector it will have a single electrical terminal the size of a normal crimp type spade connector.

If you've got a multimeter try doing a resistance check from the ECU plug's oxy sensor terminal (pin 29) to the cut wires in the engine bay. ;) It would pay to disconnect the battery before doing so though- just incase there are live 12v wires there.

Has your current oxy sensor got the proper factory style plug on it? Or is it a universal fit with just a spade cripm on it?
David
Posts: 2
Joined: Sun Jun 27, 2004 10:57 am

Post by aus.leo »

murcod wrote:Has your current oxy sensor got the proper factory style plug on it? Or is it a universal fit with just a spade cripm on it?


It is a blueish grey plug which seems to be factory style. Under the dizzy I can see a couple of wires they look like they have been ripped from a fitting.

(back in the cabin) The wiring loom colours in your pic of the ECU seem to be different colours to mine and i have three yellow plugs instead of 2. but i did locate pin 29.

The engine was Dyno Tuned running Premium ULP but my problem is the tuning on LPG, lags under acceleration.

I have since, plugged in a wire to pin 29 on the ECU and ran it through the firewall, ready to connect to the o2 Sensor.

I will test it tomorrow, hopefully in better weather.
94 CXi Feroza Petrol/LPG
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

aus.leo wrote:(back in the cabin) The wiring loom colours in your pic of the ECU seem to be different colours to mine and i have three yellow plugs instead of 2. but i did locate pin 29.

The engine was Dyno Tuned running Premium ULP but my problem is the tuning on LPG, lags under acceleration.



Be careful if your plugs are different- the ECU pin out may be different and pin 29 may not be the oxy sensor input!

If you've got the standard distributor then you probably will find the gas performance lacking. Gas requires a different timing curve to petrol for peak power- the Ferozas only run a mechanical distributor advance / retard mechanism.

Once you reconnect your oxy sensor you'll find a lot more midrange and partial throttle opening power when running on petrol.
David
Posts: 2297
Joined: Mon Mar 06, 2006 1:23 pm
Location: Melbourne-Australia

Post by MightyMouse »

A very minor addition - sensor manufacturers advise to never solder the connector onto a Lambda sensor lead - crimp only.

Most current sensors actually "breath" via the cable sheath so don't use any form of sealant on the connector

There are actually differential devices and rely on having an atmospheric reference ( hence the need to breath ) available.
( usual disclaimers )

It seemed like a much better idea when I started it than it does now.
Posts: 4760
Joined: Mon Dec 30, 2002 6:04 am
Location: Adelaide

Post by murcod »

I think most of them use stainless steel wiring- not much hope of successfully soldering that anyway.
David
Posts: 2297
Joined: Mon Mar 06, 2006 1:23 pm
Location: Melbourne-Australia

Post by MightyMouse »

Its obviously been a problem - otherwise the warning wouldn't get into the documentation. Its really the flux that is the problem, getting drawn along the sheath and polluting the sensor.
( usual disclaimers )

It seemed like a much better idea when I started it than it does now.
Posts: 1031
Joined: Sun May 01, 2005 10:22 pm
Location: Newcastle 2300

Post by joel HJ60 »

Great tech. Thank you.
[b]1985 HJ60[/b]

[url]http://forum.ih8mud.com/60-series-wagons[/url]
Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest