Boosting the 1uz/learn as you go!

[ihiryu]

Alright guys, I've been a member here for awhile now, and I've just recently boosted my 1uz. There have been a few things I've noticed that I wanted to point out to everyone that I couldn't find through searching and lurking.

A little about my car, 88 Supra, original turbo, now 1uz powered (from a SC400) w58, with the headers flipped around to a y-pipe to a holset turbo. External wastegate, 3" exhaust with no cats.

Originally the car was a 7mgte that had a BHG, I bought it and put a 2jz-geT and enjoyed it until the car cranked walked to death, but it still had no torque with only a small t04 on it.

Upset and determined I went 2jzgtte, figured the stock turbos would provide enough low end fun. It did, until number 2 turbo died it's death for being ceramic. Then I went USDM twins in true twin mode, and then it lost all it's fun.

At the time I had my SC400 which was stock (minus no cats, SAFC, and coil overs) and had a lot of fun with it, tooled around with it, and ran a little bit of NOS (since Jibby was doing it safely) and then sold the car in favor of a S2000. The Supra sat with dreams of running a 1uz with a five speed behind it, but things were getting a bit hard, and quantum auto wasn't around yet, and since I couldn't do any fab work, I could make mounts.

So I'm sure you guys know, Quantum came around, and a few other people making one off parts (like SuperRunner and his flywheels) and then going 1uz with a five speed became easy.

So fast forward, got the 1uz up and running, and I had a bunch of left over parts and a friend at a muffler shop. So I said "To Hell with it, I'm going turbo, I've gone this far to stop". And here I am today. Learning things the hard way, and wanting to pass it all down to you guys.



So far I've been running a Holset HX35 on 7 psi. I'm still on stock injectors and FPR, the only thing I've done is a Walboro fuel pump. The AFR's are still way way rich. Like 10:1 rich.

However, I think this is due to the fact I do not have o2 sensors. I don't have any plumbed in the exhaust (minus the wideband) at all so I have check engine lights. I've also noticed that the adjustments I have made with my SAFC have NOT been relearned. So I think I've combated the factory ecu's relearn function.

To test whether the ecu is relearning or not, I've leaned out the car to about 16:1 at low throttle at 3500 rpms'. I've logged about 300 miles without any relearn yet. So far so good! On my stock SC400 ANY changes I made with the SAFC would be relearned in a matter of days, would take about a 100 miles or so.

In my meager 300 miles of driving I blew out my cam seals because of crank case pressure. On the left bank (driver side), I replaced the PCV with one from a Supra TT, and a catch can. However I didn't realize the right bank (passenger side) with the hose going into the throttle body (even though Lex and XR mentioned it) was still connected.

What I plan to do is vent that to atmosphere for now and see what will happen.

So far so good. I've got the KMV/AFM sensor blow through in the stock location with aluminum piping. I know I have a leak through the little filter adapter, but no big deal. I had to come up with a way to do an oil return since I couldn't find a way to gravity feed it so I used a pump to do it.

But that's about it, once I find some more issues I'll definitely post back up.

 

[stevechumo]

I was told you could use the O2 sensor simulators to trick the ECU to think everything is normal. I see there're a few of them on Ebay.

 

[ihiryu]

I meant the primary front o2 sensors

 

[cribbj]

Some good practical lessons there; w/o the O2 sensors the ECU is probably running in open loop limp mode, and that's why it's not relearning. It would be interesting to know what other functionality isn't working when in limp mode - probably timing isn't being adjusted, and is preset by the ECU to some very conservative figure?

 

[XR8tt]

Yep !! Which is possibly so close to whats required..
I wouldn't push it too much further...
Without upgrading fuel supply...

Aha we agree Cribbj ?? Must be wrong then ?? Haha....

 

[cribbj]

Aha we agree Cribbj ?? Must be wrong then ?? Haha....

Graeme, I think the only times we disagree is when you're wrong about something

 

[ihiryu]

Lol,

Okay I got my 315cc injectors in, and I pulled -20 from 500-1500 RPM's. Then from 2000-redline I pulled -25 on the SAFC. The car ran very very good at partial throttle.

On the high map I pulled -10 across the board. I hate to say it, but the car feels a lot faster now. I hope it's just not the butt dyno. Right now, I'm about 11:1. It's weird cause before I was at 10:1 with the stock injectors. I'm going more I can pull to get it to 12:1 under boost.

I've got my throttle set at 15% and 40%.

*disclaimer* these are my settings, it may not work for you, and might blow your motor.

To my understanding the only way to check timing while a car is running is only via a dyno and a timing light correct? Is there some other way to monitor it?

 

[ihiryu]

So I was finally able to compare my wideband to another, and had to get a new WB. So got my new AEM Bosch sensor in, and tried it again.

Whoo, good thing I didn't beat on it too much. My AFR's were way out of whack, so let this be a lesson to everyone, make sure you get your wideband checked!

Anyway, so I have some really really weird settings right now on the SAFC. At idle, I had to leave it at zero otherwise it would idle really really bad. Off the top of my head,

Lo (in 500 increments starting at 500) 0, 0, -25, -25,-25 (across the board)
Hi (in 500 increments starting at 500) 0, 0, 0, +30, +30, +0 that's as far as I've gotten. At about 3K and up it goes super super rich at WOT. Like the 10:1 area

I had to change my Thr settings to 15/50 to get drivability better. I do get 15.5:1 AFR's at low throttle now which is perfectly fine for street driving. The car doesn't miss a beat at this low of an AFR. However, when you go to WOT, you can feel a hard shudder from the transition of 15:1 to 11:1.

Also, I need someone else to test this, but when I made a change at WOT, it took a few pulls BEFORE the change would take effect. Like for instance, I'd go WOT and it would go lean at 3K, and this was at my first initial setting at +25, and it would stay lean, like 14:1 under WOT lean. I then adjusted to +40 and tried again and it still stayed lean. So fed up, I adjusted to +50 (max on the AFC), and still no effect.

I left it there, and kept doing WOT pulls, just to verify that it was going lean at the 2900-3K area. I was starting to think the fuel pump, or the fuel pump circuit wasn't supplying enough voltage to the pump. Then suddenly after four or five pulls, it went rich, 10:1, then I started to back it out, and it got leaner, and finally now it's sitting at a pretty 11:1 at WOT.

This was all last night, so this morning I went for my hour long ride back to work, I went WOT, and it still read 11:1 at WOT at 25-3500 RPM. So it seems like it took a few pulls before the AFC settings took effect on the car.

I'm still going to be driving and testing it to make sure that the car doesn't relearn. I can't say for sure whether or not the ECU has been trying to relearn since my last o2 sensor was apparently FUBAR'd.

Of course, I'll report back in a few days on my findings. The only issue I have without running primary o2 sensors is cold starts. When I first crank it up, I've got to rev it for a few seconds before she'll idle on her own. But after that she'll run great unless I blow off and she'll die. But that could be from; lack of decel air settings, or eliminating the dash pot. I'd say both just because my BOV is AFTER the AFM. The reason was because I didn't have enough room before the MAF to put it. Oh well, it only happens when it's cold outside so I can definitely deal with it.

 

[stevechumo]

The BOV should be at least 12" before the MAF. The easy way to avoid the ECU to relearn is to use the 02 sensor simulators. I talked to the engineer at Split Second and they also recommended.

 

[ihiryu]

Could you direct me to these o2 sims? I don't think I've ever seen one for a four wire o2 sensor.

 

[gloverman]

Now I know why I try to fit two wideband units to the car im tuning. Next time I have a stock ECU 1uz on the dyno Ill unplug the o2 sensors and see if the timing changes. I suspect that only the fuel map alters with them unplugged and then of coarse no relearn as it needs the o2 sensors to learn. Whch model of ECU is in your car? Ill try and use the same one. Cheers

 

[stevechumo]

Could you direct me to these o2 sims? I don't think I've ever seen one for a four wire o2 sensor.

Now, I only see there're 2 on Ebay, but not for the US market. However, i guess they can be used because they're universal with 4 wires.

http://cgi.ebay.com/Buddy-club-02-S...arts_Vehicles_CarParts_SM&hash=item3a5fad212d

I used to see a lot of them on Ebay last year, but not anymore. Sorry, you'll have to google it.

 

[ihiryu]

I'm using a 92 SC400.

AEM's Wideband doesn't support narrowband output. I know with the zeitronix unit you can simulate it, and in the programing you can actually force what signal you want the ecu to see.

So for instance, you can force the ecu to see the 14.7 even if you're really running 12:1 or 15:1, but I wonder what timing changes the ecu would see based on AFR's alone.