1JZ boosts horribly when cold

[Rich]

I think it's the O2 sensor, the easiest way is trying one from a car that is working properly.

do you have an onboard wideband? when warmed up and in closed loop situations you should be able to see the a/f's bouncing up and down between 14.2 and 15.2 (roughly) at a pace of two to three times a second (the 'cross counts')

you can also check that with a volt meter which is able to react quickly enough to voltage variations, it should vary around 0.45V

 

[88mkIIIproject]

i don't have a wideband, nor an accurate volt meter...

can i install a wideband 02 as an oem replacement, without it hooked up to anything else?? wasn't sure how that worked by itself.

i DO need a good voltmeter, maybe now is the time...

thanks

 

[Tim]

its not a fault in the engine. Every performance high horsepower turbo car has done the same exact thing in my experience. You must give the car time to reach operating temp, and then its fine. How can it be an o2 sensor if the car is running in open loop when its warming up, and its not using the o2 sensor? Every turbo motor that I have owned rb20det, 1jzgte, (3) 4g63t's, b16a-t, has run like crap unless I let it warm up for 5-10min.

 

[Rich]

first few seconds after start it is not running in closed loop, you are correct in that, but within 20 to 30 seconds it will start running in closed loop and start searching for the 14.7 a/f

@88mkIIIproject, no, you cannot install a wideband as an oem replacement, they work in a completely different way. You will need to have the exact same narrowband O2 sensor for replacement.

A wideband is for monitoring purpose only, it does not work with our stock ecu's or blitz/mines modified ecu's.
Standalones can run of a wideband though.

 

[rakkasan]

A wideband is for monitoring purpose only, it does not work with our stock ecu's or blitz/mines modified ecu's.

Not true, I ran a LC-1 as my narrow band with good success. Aaron hooked it up to the stock ECU with two of the wires & ran another set of wires into my Profec E-01 so I could have real time AFR reading....

 

[Rich]

let me explain
The narrowband which is used on our 1jz's (zirconia type) generates a voltage based on the amount of oxigen in the exhaust gasses. This narrowbands cross count voltage is 0.45V and works binary: The ecu looks at the 0.45V, anything below 0.45V it reads 'leaner than 14.7' anything higher up to 1.1V (which is the max voltage a narrowband can produce) it reads as 'richer than 14.7'. How lean or how rich our ecu's can't read, it only works with 'below' or 'above' stoich.
This only applies to closed loop situations.

A wideband has a range of 0 - 5V (where 0V is all air and 5V is pig rich) but doesn't generate a voltage by itself, it needs a reference voltage (that's the fifth wire) from a controller or ecu.

So, please tell me how a wideband, missing it's reference voltage and using a scale of 0 - 5V works on an ecu which expects to see a voltage varying around 0.45V?

 

[rakkasan]

let me explain
The narrowband which is used on our 1jz's (zirconia type) generates a voltage based on the amount of oxigen in the exhaust gasses. This narrowbands cross count voltage is 0.45V and works binary: The ecu looks at the 0.45V, anything below 0.45V it reads 'leaner than 14.7' anything higher up to 1.1V (which is the max voltage a narrowband can produce) it reads as 'richer than 14.7'. How lean or how rich our ecu's can't read, it only works with 'below' or 'above' stoich.
This only applies to closed loop situations.

A wideband has a range of 0 - 5V (where 0V is all air and 5V is pig rich) but doesn't generate a voltage by itself, it needs a reference voltage (that's the fifth wire) from a controller or ecu.

So, please tell me how a wideband, missing it's reference voltage and using a scale of 0 - 5V works on an ecu which expects to see a voltage varying around 0.45V?

Dunno, ask Aaron. My 1J came back from CA without a Ox sensor, only a LC-1 in the downpipe. It does work.

 

[dugums]

let me explain
The narrowband which is used on our 1jz's (zirconia type) generates a voltage based on the amount of oxigen in the exhaust gasses. This narrowbands cross count voltage is 0.45V and works binary: The ecu looks at the 0.45V, anything below 0.45V it reads 'leaner than 14.7' anything higher up to 1.1V (which is the max voltage a narrowband can produce) it reads as 'richer than 14.7'. How lean or how rich our ecu's can't read, it only works with 'below' or 'above' stoich.
This only applies to closed loop situations.

A wideband has a range of 0 - 5V (where 0V is all air and 5V is pig rich) but doesn't generate a voltage by itself, it needs a reference voltage (that's the fifth wire) from a controller or ecu.

So, please tell me how a wideband, missing it's reference voltage and using a scale of 0 - 5V works on an ecu which expects to see a voltage varying around 0.45V?

My wideband features a narrowband simulation that can connect to the factory ecu, though I am not using it. (Zeitronix )

 

[Tim]

Every car I have seen will stay in open loop and run on the rich side with set values until the coolent temp sensor tells the computer that the engine is at operating temp, then the car will go into closed loop and the o2 sensor will start to cycle rich/lean/rich/lean to try to hit that 14.7 stoich A/F ratio. I work at Nissan and when the engine is cold you have to hold the car at 2000 rpm for a few minuts to build heat in the engine before you can see the o2 start to cycle and go into closed loop. If you ever see a car with a bad coolent temp sensor, it will run pig rich and not go into closed loop.

 

[88mkIIIproject]

My wideband features a narrowband simulation that can connect to the factory ecu, though I am not using it. (Zeitronix )

do you know any other details on how this works??

 

[dugums]

http://www.zeitronix.com/installation/ZT2instructions.pdf
http://www.zeitronix.com/questions/NBpoint.htm

 

[c_ronius]

The lc-1 has a narrowband simulation output as well... that's how the stock ecu uses it. It has to be a feature of the wideband controller.

 

[88mkIIIproject]

http://www.zeitronix.com/installation/ZT2instructions.pdf
http://www.zeitronix.com/questions/NBpoint.htm

thats awesome, thanks man

 

[Rich]

Every car I have seen will stay in open loop and run on the rich side with set values until the coolent temp sensor tells the computer that the engine is at operating temp, then the car will go into closed loop and the o2 sensor will start to cycle rich/lean/rich/lean to try to hit that 14.7 stoich A/F ratio. I work at Nissan and when the engine is cold you have to hold the car at 2000 rpm for a few minuts to build heat in the engine before you can see the o2 start to cycle and go into closed loop. If you ever see a car with a bad coolent temp sensor, it will run pig rich and not go into closed loop.

I have cold started my engine several times with the eman blue monitoring on a laptop to tune my coldstart/idling. I have an onboard aem uego wideband hooked up to the eman.
On my engine it starts going into closed loop within 30sec, even with these winter temps. I assume all 1jz's using the same ecu do the same, as for nissan I wouldn't know, not really relevant on a toyota forum...