Turbo O2 sensor?

[presslab]

Does anyone know what type of O2 sensor the turbos use?

 

I'm trying to track down my surging at mid-heavy throttle problem. The "power waves." I have been using a regular Bosch universal sensor (Zerconia) but I think the problem is that it doesn't respond fast enough. I have the original sensor and the voltage swing is much greater than the Bosch one. I think it is a Titania type versus the Zerconia.

 

The Titania are not very common, the Bosch web site says they are used in 0.5% of all cars. The Titania are also much faster responding than the Zerconia ones.

 

Maybe this is the source of the infamous turbo surging problem, I dunno. I am running the original (old) sensor and so far the surging is gone. The problem is intermittent so time will tell. I'd still like to get a new sensor though, but the price for the exact replacement one is a lot.

 

 

I also think I can come up with a way to easily use the 260cc injectors with the stock engine management system. I have figured out how to modify the hot wire MAF to adjust the output signal. My engine was so very rich at WOT so I leaned it out by adjusing the MAF. It actually pulls hard all the way to redline where as before the power dropped off. I hope for better mileage too.

 

It involves opening up the MAF and adding a variable resistor. I ordered the 260cc injectors and I'll have a go at making it work. I'm not sure how the fuel map and timing map will work out.

 

If the 260cc injectors work then next is a simple PWM circuit to drive the boost duty solenoid. I plan on 11 lbs of boost.

[joostvdw]

 

I also think I can come up with a way to easily use the 260cc injectors with the stock engine management system. I have figured out how to modify the hot wire MAF to adjust the output signal. My engine was so very rich at WOT so I leaned it out by adjusing the MAF. It actually pulls hard all the way to redline where as before the power dropped off. I hope for better mileage too.

 

It involves opening up the MAF and adding a variable resistor. I ordered the 260cc injectors and I'll have a go at making it work. I'm not sure how the fuel map and timing map will work out.

 

If the 260cc injectors work then next is a simple PWM circuit to drive the boost duty solenoid. I plan on 11 lbs of boost.

 

sorry, can't help you with the O2 sensor, but I'm interested in your MAF shenanigens :-p

 

You want the MAF to give a lower signal than normal, so you can adjust for the increased flow of the injectors, so why not just mount an variable resistor on the outside of the maf sensor? Or go beserk and use an op-amp with negative amplification

 

With the boost increase, you can just chuck the boost solinoid and use a boost controller, be it manual of AVC-R style. On the other hand, if you're into electronics, why not make your own, just unhook the original boost solinoid so the ECU won't interfere with it.

[GeneralDisorder]

You could use a wideband sensor that can mimick the output of a narrow-band sensor. Many of the wideband kits availible can do this. I would think that should be about the best narrow-band output you could get and you have the added benefit of a wideband for your other tuning needs.

 

I got the wideband from NGK - I haven't installed it yet but I like what I've seen of it so far (sitting in the box still ). It has the narrow-band output so you don't have to add a second O2 bung to cars that require a narrow-band input to their ECU.

 

GD

[presslab]

sorry, can't help you with the O2 sensor, but I'm interested in your MAF shenanigens :-p

 

You want the MAF to give a lower signal than normal, so you can adjust for the increased flow of the injectors, so why not just mount an variable resistor on the outside of the maf sensor? Or go beserk and use an op-amp with negative amplification

 

With the boost increase, you can just chuck the boost solinoid and use a boost controller, be it manual of AVC-R style. On the other hand, if you're into electronics, why not make your own, just unhook the original boost solinoid so the ECU won't interfere with it.

 

The problem is that the MAF will (probably) clip the output signal with more air flow when I turn up the boost. That's why modding internally should have the ideal result. And I didn't have to add an external circuit with all that entails. Just cut one trace and solder a resistor (variable or not) in. I'll try to upload a pic later.

 

I don't really like manual boost controllers... They probably work OK but the boost curve they have is not the best for smooth power at part throttle. If I just apply some duty cycle to the boost solenoid it will just make the wastegate see that average, right? Like at 50% duty cycle the wastegate would see half the boost pressure I think. Is my theory flawed on this?

[presslab]

You could use a wideband sensor that can mimick the output of a narrow-band sensor. Many of the wideband kits availible can do this. I would think that should be about the best narrow-band output you could get and you have the added benefit of a wideband for your other tuning needs.

 

I got the wideband from NGK - I haven't installed it yet but I like what I've seen of it so far (sitting in the box still ). It has the narrow-band output so you don't have to add a second O2 bung to cars that require a narrow-band input to their ECU.

 

GD

 

You know I have a wideband sensor in the same spot, on the shelf. I haven't made up the necessary circuit to get it to work though. I wonder what the response speed of the wideband sensor is? I think they are Zirconia too?

[ausubaru92]

Your O2 sensor shouldnt make anydifference when you are on boost. At about 1 psi, the ecu goes into open loop mode and ignores the O2 sensor.

 

A bosch universal one is fine. Im using one.

[[HTi]Johnson]

Your O2 sensor shouldnt make anydifference when you are on boost. At about 1 psi, the ecu goes into open loop mode and ignores the O2 sensor.

 

A bosch universal one is fine. Im using one.

 

I was thinking the same thing, but if he has the stock one in and it doesn't surge anymore...wtf.

[presslab]

Your O2 sensor shouldnt make anydifference when you are on boost. At about 1 psi, the ecu goes into open loop mode and ignores the O2 sensor.

 

A bosch universal one is fine. Im using one.

 

Mine doesn't do that. I have an O2 sensor "gauge" thing and it still uses the O2 sensor at ~7 PSI boost. (Meaning the "gauge" flickers back and forth.) At full throttle it does not use the O2 sensor.

 

To diagnose this problem I unplugged the O2 sensor and the surging went away too. I think the older turbos did like you said because the service manual mentions something like that. I have an 88 with hot wire MAF and the boost switch only controls the dash "turbo" light. It does not know when it is under boost it only knows the air flowing into the engine and throttle position.

[presslab]

Here is the pic of the mod to the MAF. I am using a 100 ohm potentiometer and right now (with stock injectors) it is at 60.5 ohms.

 

You can see the cut trace in the upper right corner.

[presslab]

Ok I found in a catalog on the internet that my sensor is a Titania one. It looks like only the California turbo models had it, which my car is. I found a new sensor on sparkplugs.com for $59 so I ordered it.

[presslab]

Put the new o2 sensor in this morning, it's definitely a Titania one. The Titania use a flat strip instead of a cone shape. Attached is a pic of the old sensor, I cut it open.

 

http://www.tomco-inc.com/Tech_Tips/ttt20.pdf

 

 

I have also installed my brown top 280ZX Turbo injectors I bought from Dr. Injector. I tuned the MAF to offset the additional fuel.

 

So far the car is running great! Next step is a bit more boost.

[joostvdw]

Put the new o2 sensor in this morning, it's definitely a Titania one. The Titania use a flat strip instead of a cone shape. Attached is a pic of the old sensor, I cut it open.

 

http://www.tomco-inc.com/Tech_Tips/ttt20.pdf

 

 

I have also installed my brown top 280ZX Turbo injectors I bought from Dr. Injector. I tuned the MAF to offset the additional fuel.

 

So far the car is running great! Next step is a bit more boost.

 

sounds good, do you have an wide band sensor to check your fuel injection? It sounds like you may have found a way for us to use bigger injectors (which means more boost more power and all the good stuff )

 

keep us updated!

[presslab]

sounds good, do you have an wide band sensor to check your fuel injection? It sounds like you may have found a way for us to use bigger injectors (which means more boost more power and all the good stuff )

 

keep us updated!

 

Well I do have the wide band sensor itself but I never made the circuit, so no I don't really have one. I am just looking at the voltage of the narrowband O2 sensor. I adjust it until it's basically stoich and then make it a bit richer.

 

A lot of people say the stock injectors are 180cc, but at what pressure? The 280ZXT are 260cc @ 43psi. Someone on here measured the stock XT4T at 210cc @ 43psi, which I think my GL10 has the same injectors.. I mention this because I didn't really have to change the MAF that much with the new injectors.

[presslab]

I made up a little circuit to send a PWM signal to the wastegate duty solenoid, for more boost. It has a dummy load resistor so the ECU does not throw a code. The original signal from the ECU is a 20 us pulse at 15 Hz, basically 0% duty. However with this short pulse it can detect if the solenoid is plugged in and throw a code if not. I read somewhere that the ECU will up the boost at high altitude with this signal, but I am at sea level. A 50% duty is giving me about 10 PSI.

 

Unfortunately my engine is missing at high boost + high RPM. The O2 never goes lean but I'm pretty sure one cylinder is fouling out because when it starts missing and I back off the throttle, the engine is lumpy for about 30 seconds and then smooths out again. A flooded cylinder I guess.

 

In trying to figure that out I have played with a variety of MAF adjustments. I can get it to fuel cut at ~9 PSI, or if I adjust it "lean" it can do ~15 PSI easy. The injectors still flow enough to not actually go lean though. I think the MAF adjustment + bigger injectors is going to be fine.

 

I have on order new plugs, wires, cap, and rotor. We'll see if that fixes my engine missing.

[ple1ades]

So I have a EA82t no rear o2 sensor is that why I am running rich as hell?

[Turbone]

I made up a little circuit to send a PWM signal to the wastegate duty solenoid, for more boost. It has a dummy load resistor so the ECU does not throw a code. The original signal from the ECU is a 20 us pulse at 15 Hz, basically 0% duty. However with this short pulse it can detect if the solenoid is plugged in and throw a code if not. I read somewhere that the ECU will up the boost at high altitude with this signal, but I am at sea level. A 50% duty is giving me about 10 PSI.

 

Unfortunately my engine is missing at high boost + high RPM. The O2 never goes lean but I'm pretty sure one cylinder is fouling out because when it starts missing and I back off the throttle, the engine is lumpy for about 30 seconds and then smooths out again. A flooded cylinder I guess.

 

In trying to figure that out I have played with a variety of MAF adjustments. I can get it to fuel cut at ~9 PSI, or if I adjust it "lean" it can do ~15 PSI easy. The injectors still flow enough to not actually go lean though. I think the MAF adjustment + bigger injectors is going to be fine.

 

I have on order new plugs, wires, cap, and rotor. We'll see if that fixes my engine missing.

 

What about your timing? Have you advanced it at all?

Or did I miss that

[presslab]

So I have a EA82t no rear o2 sensor is that why I am running rich as hell?

 

There is only one O2 sensor on these cars... So you have none at all? Without the O2 sensor it can run very rich, the computer doesn't know the mixture and if your MAF is off it can be bad.

[presslab]

What about your timing? Have you advanced it at all?

Or did I miss that

 

20 BTDC w/ green plugs together, it's at stock timing. 91 (R+M)/2 octane gas.

 

So far I've put on a MSD 8222 coil which did help (BTW the stock coil is 0.8 ohm and the MSD is 0.7 ohm) and new NGK BPR6EY-11 plugs which helped too. Gapped plugs to 0.030". Plugs look good, nice even tan color. Compression is from 122 to 126 PSI across all cylinders.

 

Still misses though at high boost. The exhaust note changes notably to an uneven note. I let off for a bit and then get back on it and sometimes it will smooth out and pull away. Waiting for the wires and cap/rotor to arrive.

[ausubaru92]

I made up a little circuit to send a PWM signal to the wastegate duty solenoid, for more boost. It has a dummy load resistor so the ECU does not throw a code. The original signal from the ECU is a 20 us pulse at 15 Hz, basically 0% duty. However with this short pulse it can detect if the solenoid is plugged in and throw a code if not. I read somewhere that the ECU will up the boost at high altitude with this signal, but I am at sea level. A 50% duty is giving me about 10 PSI.

 

Can you give some details on this circuit, im very interested.

 

So I have a EA82t no rear o2 sensor is that why I am running rich as hell?

 

What do you mean rear? some turbo models had an exhaust temp sensor at the bottom fof the cat.

 

You will run rich at idle and cruise if you dont have an o2 sensor, but when boosting above about 2psi, the engine goes into 'open loop' mode and ignores the o2 sensor

[4WDFrenzy]

presslab, you are awesome! That is so cool that you are figuring this stuff out. This is a problem that has plagued most of us for awhile now. Definitely keep us updated. I am very interested in seeing how your build turns out(and no I'm not sitting back waiting for the whole thing to blow either). I really think that you might have the key to unlocking some secrets(on the electronic side) of these engines. Good Work & Congrats! You are truly an inspiration!

 

Patrick

[presslab]

Can you give some details on this circuit, im very interested.

 

Here is the schematic. It is basically a 555 timer that generates a variable duty cycle PWM signal. The frequency is around 15 Hz.

 

It works pretty good. On shifting (automatic) I get maybe 1 PSI spike. It holds boost to redline; it's way better than the MBC I tried.

[presslab]

presslab, you are awesome! That is so cool that you are figuring this stuff out. This is a problem that has plagued most of us for awhile now. Definitely keep us updated. I am very interested in seeing how your build turns out(and no I'm not sitting back waiting for the whole thing to blow either). I really think that you might have the key to unlocking some secrets(on the electronic side) of these engines. Good Work & Congrats! You are truly an inspiration!

 

Patrick

 

Thanks! I'll keep you guys updated on my progress. I think I'm getting pretty close.

[joostvdw]

it IS looking really good indeed, good job

 

One thing has me puzzled though, this boost solinoid is constantly opening and closing right? don't you think this will fail shortly if you keep running it like this? it was not designed to work like this for prolonged periods right?

[presslab]

it IS looking really good indeed, good job

 

One thing has me puzzled though, this boost solinoid is constantly opening and closing right? don't you think this will fail shortly if you keep running it like this? it was not designed to work like this for prolonged periods right?

 

Yeah, you might be right. I thought about that before as my solenoid is clicking away at idle; it's not that loud though. My thought was to connect my circuit to the "Boost" light on the dash, so it is only working under boost, but I haven't got around to that.

 

edit: I attached the ground of my circuit to the boost switch, now the solenoid only acts under boost.

[presslab]

Last night I drove the car in low 30s F temps and had some surging. It was under ~5 PSI boost, part throttle, and I could see the ECU still trying to close the loop.

I have found the ESC1 product which disables the O2 sensor under boost:

http://www.splitsec.com/products/conditioner.htm

From reading it seems like people use it when they add turbos to a non-turbo engine.

The ECU only knows the air mass into the engine. By tweaking the MAF I have fooled it into thinking there is less air, and the ECU is keeping the loop closed under more load. However I have noted the closed loop under boost with even the stock injectors and MAF. The stock ECU seems a little aggressive trying to close the loop. Perhaps the later ECUs didn't do this? Mine is an '88 GL-10.

So I think I will try adding a relay to the boost switch. I will wire it to disconnect the O2 sensor under boost. Hopefully that will send the ECU into open loop mode. My guess is that's all the ESC1 does.