3rd Annual Subaru Shootout: Recap: MAF discussion

[BAC5.2]

Gotta say, this was one of the best weekends I've had in a while. Fast cars, good friends, in-a-fix repairs. Good times.

It was great to finally put a few more names to faces and meet you guys.

Matt and I seemed to be vying for fastest Turbo Legacy. Despite his engine troubles, he edged me out with a 15.5 @ 15psi (I think I ran a 16 flat @ 11psi fuel cutting at the top of 3rd, and essentially coasting for the last 60 feet. But Matt blew his hose at about the same time.)

My best run of the day was a dismal 15.601 @ like 88mph (@10psi). Highest trap was 90.76mph, best 60' was like 2.401, and best R/T was .441. I need to listen to Ben's and Matt's advice about launching, and hopefully drop those 60' times down a few tenths.

Again, I really had a blast out there with you guys.

Corsair - If you need those videos hosted, you can send them to me. Just instant message me or something tomorrow night and send them that way.

I can actually host everyone's pictures (and when I say "I" I really mean Hardy can ). The only pictures I took were of Ben's and my UBER-Ghetto Intercooler fix for Vikash. (he tore a hose so we stuck a radiator hose I had left over from the 16G install and put it on Sam's (Corsair) exhaust for a few minutes to warm it up. Then just forced it onto my spare all-aluminum IC and Ben wrestled with getting the turbo pipe back on). It was surprisingly fun, and I'm really really stoked that we could safely get Vikash home (and even take a few more runs).

Oh, a Forester 2.5XT had a R/T of .004. I've never heard or seen anything like that before in my life.

That's pretty much it. We just mostly talked and shot the shit. There were 3 Turbo Legacy's (myself, Matt, and Vikash) 2 N/A wagons (with us), an EJ22T powered Impreza (who got me out of the hole by about .1), multiple WRX's and STi's too. On the drive from the meeting point to the track we had 11 Suby's rollin along, then Corsair met up with us and was our 12th. It was a hell of a sight in my rear view mirror, to see a whole friggin TON of Subaru's cruising in.

Anyway, it's 12:37AM, and I've gotten like 4 hours of sleep in the past 48 hours, so I've rambled on to long.

[dzx]

What problems is matt having with his engine and what kind of fuel cut did you run into? I was reading launching techniques that guys use on their wrx and they must have more low end torque cause it didn't work as well for me.

[NemesisEJ22t]

At this track, the launching was particularly hard, I think that had quite a bit to do with the times they were getting. The staging area was on an angle so that the cars would roll back while you were waiting for the green light. Other than that, it was a blast. It was especially good to get away from school for a day. We do need to set up something in the winter so that i can bring my Legacy out though .

[Tleg93]

Are you guys going to post your times?

[BAC5.2]

Yea, I was sitting at the line jockeying the e-brake so I could stage correctly. And being my first time ever launching the car, I was having... issues, lol. One guy in a wrx pulled 1.95 second 60' times! He also made like 50 runs, lol.

dzx - I think we can get the WRX downlow, it's up top that we suffer the DOHC fury. I'll let Matt talk about his car.

Don't know what kind of fuel cut. Certaintly not Overboost fuel cut, so it's something else.

[legacy92ej22t]

I'll let Matt talk about his car.

I had a great time but it was a bit of a sad day for me. My worst fears were confirmed at about 4:30 am Saturday morning when Vikash, Phil and I finished doing a compression test on my engine. I had been getting steady power loss over the week leading up to the Shootout and suspected either a burnt valve or that I toasted a ring. I got 158-160 psi on 1-2-3 but only 72 psi on #4. Put a little oil in the cyclinder and it shot up to 170 psi. Looks like bad rings at best and a cracked piston and gouged cylinder walls at worst. So anyways, we finally got to bed at 5 am, got up at 7:30 and went to the shootout and I raced anyways on 3.5 cyclinders. I never got a solid run in because I was having intermittant fuel cut issues and my compressor outlet to TMIC hose kept blowing off. I could only manage a 15.5 at 92 mph. My hose blew off about 200' before the 1/4 on that run too and I just coasted across the line. My best R/T was a .134 IIRC with a 60' of 2.1. Pretty good considering that I had no torque because of the lack of compression on #4. I had high hopes but alas, it wasn't meant to be this year. I'm looking at a rebuild now (I hope) so hopefully next year will go a bit smoother.

Thanks to Matt and all the W&L folks for again having the Shootout, and thanks to SOA and TXS for showing up too. It was great time even though I was zombie tired and felt like rolling into a ball and just crying myself to sleep. Uh, I think I'll just go do that now.....

[BAC5.2]

Aww Matt, she'll be OK. Just have faith

I wish Jeff and Jon had shown up. I really wanted them to see my progress as well (and talk to them about an uppipe).

[evolutionmovement]

Yeah, I found bits of my valve stuck to the screen in the pan (lots of bits - you guys saw how much melted away) and a few in the cylinder, but none of those pieces damaged the cylinder wall. Even the valve seat is fine. If I put a new exhaust valve in, it would've done the trick with no other work but a thorough cleaning. So anyway, I'd think positive that the damage is isolated to one component. Hm, maybe time to upgrade the pistons? Fueling, I bet, was the issue.

How long before you take the engine apart and what are you going to use in the meantime?

Steve

[boostjunkie]

Phil, you need to get a fuel pump put in and raise the boost on that sucker! 10psi is very close to being out of the efficiency range for those turbos.

[vrg3]

Yeah, this was definitely one of the best weekends in recent memory for me too. It was great.

My personal first experience racing was kind of mixed. At first I was doing really badly, and realized that I was doing very conservative street launches. Then I started to try being more aggressive with the clutch, and I just kept bogging off the line. I couldn't make myself be hard on my clutch!

So it was established: I suck at launching. My 60-foot times were in the mid-3-second range. So I kept doing passes trying to learn to launch, but I couldn't manage to do it well. I was steadily improving but I was getting pretty close to the limit of how aggressive a launch could be without my clutch losing grip. I suspect my clutch is pretty far worn... I know that the pedal feels way softer than Subarina's (who actually has more mileage than mine on the stock clutch). Anyhow, the best 60-foot time I managed was 2.701 seconds which gave me my best ET: 17.420 seconds at 77.31 mph.

One of the next passes was interrupted by a loud pssshhhhhhhhhhheeeeeeeeeww sound. One of my intercooler hoses had finally given up. It was a rubber sewer drain pipe, not rated for temperature, pressure, or oil, so I guess it was only a matter of time anyway. It sucked though because I didn't have any other hose on hand and I didn't want to have to quit racing for the day and have to limp my car from Williamsport back to Ithaca afterwards.

The next part was one of the best parts of the day. As people started to realize what had happened, they all came over to help me figure out how to fix it. Phil handed me a piece of radiator hose he had (I think Matt actually gave it to him originally) and I tried to make it fit over the intercooler inlet. It wouldn't; it's OD was smaller than the intercooler fitting's OD.

I gave up on it temporarily because Matt came back from a pass saying he'd hit fuel cut during the run. I decided to ride shotgun for a few passes and watch the scan tool while he raced (more on this later). Every time we passed by my car, there were more and more guys crouched over my hood trying to make it work! I was impressed and extremely appreciative of all the help everyone gave me. They were rooting through my trunk for parts... They pulled out a funnel that might stretch the hose, and then determined that the way to go was to stretch the hose a little and then fit it over a warm tailpipe to help make it hold that size long enough to force over a spare all-metal Saab intercooler Phil had brought (it would have crushed a plastic intercooler inlet). Next thing I know, they've got the hose on and all I had to do was cut it to the right length, and Phil wrestled it into place for me.

Thank you to everyone who helped me with that. I hope I'm not forgetting anyone, but thanks to Phil, Ben, Sam, and ... the guy with the black WRX who swapped the EJ22T into his girlfriend's Impreza (sorry I forgot your name!) for getting my car running again. I love the old school crowd.

After that I backed the boost down to the wastegate level -- 5 psi -- and decided to just keep practicing my 60-foot times. I never did better than that 2.701 seconds but I think my clutch was a little glazed.

One really fun thing that happened, though -- someone (I don't think he's a board member) came with his 4EAT NA wagon, and entered it when they dropped the entry fee to $10 towards the end of the day. We raced after I had my boost turned down. He and I were nearly dead on at the end of the 1/4 mile! Both times we were like within hundredths of a second of each other. He'd get away from me off the line and then I'd gain on him and pass him right at the finish line. Those two raced (one I won, one he won) were the most fun passes I did all day just because of how close they were. It really doesn't matter if you're slow if you're also racing someone slow.

The night before wasn't so great... The party was fun at W&L, and it was cool to get some free stickers and carabiners and to hang out with people and to see under the hoods of newer cars that I wasn't familiar with, but when we got back to Matt's and did the compression test it was sobering.

Earlier in the day we'd discovered the intercooler hose was leaking so we replaced the muffler clamps with hose clamps, and we Seafoamed the intake, and the car felt a good deal better than it had, but we weren't sure whether or not it was 100%. I couldn't believe the results of the compression test. How does an engine with 50% compression on one of its cylinders run so smoothly? And how does damage like this happen to an EJ22T? Subarina will eventually be okay, but this really sucks.

About this fuel cut thing -- I've learned a lot about our ECUs this past week. Both Phil and Matt have been hitting fuel cut when driving really hard. And it's not overboost fuel cut; we have verified with the scan tool that the ECU isn't seeing the overboost condition. And it's not misfire; the ECU is definitely cutting fuel.

But there is something else. I have to verify this with a ghetto flow bench, but I believe that the stock EJ22T MAF sensor is a severely limiting factor in power improvements. If I've done my analysis right, its range only goes up to about 200 grams per second, or roughly 200 horsepower's worth of air. 200hp. That's it. The MAF sensor's range is exceeded before the injectors' capabilities! Our best guess is that when the ECU sees the sensor hit the limit it just throws the injectors open at 100% duty cycle, since it's the best guess.

So it's possible that this fuel cut is some kind of rebellion against seeing the MAF sensor at its rail for too long, since that's certainly an abnormal condition. But Jason, you never experienced fuel cut problems after installing my FCD, right? So I don't know what to think... it's true that the code in Phil's and Matt's ECUs is different from the code in the boostjunkie Legacy's ECU, so maybe something changed. I don't know. But it's clear that this is something that needs to be addressed.

So anyway, I wish we'd taken the opportunity to take more photographs. I know corsair got some, and also some videos (I think there's one of me against an STi, and it was hard to keep both of us in the frame ). Oh well, more next time.

Thanks, Matt, for your hospitality and for taking charge of the Legacy Central constituency. Thanks to W&L Subaru for hosting the whole event. Thanks to TurboXS for the prizes raffled off. And thanks to everyone who showed up and made it so much fun.

[evolutionmovement]

Even with a melted exhaust valve for a few miles at 80 the only way I could tell anything was wrong was the weird exhaust note. Then it started running like a bad ignition lead.

If the MAF is a big detriment, is that the sensor's range or the ECU's? How do people run big numbers with aftermarket ECUs if it's the MAF's limitation?

Steve

[boostjunkie]

[quote="vrg3"]But there is something else. I have to verify this with a ghetto flow bench, but I believe that the stock EJ22T MAF sensor is a severely limiting factor in power improvements. If I've done my analysis right, its range only goes up to about 200 grams per second, or roughly 200 horsepower's worth of air. 200hp. That's it. The MAF sensor's range is exceeded before the injectors' capabilities! Our best guess is that when the ECU sees the sensor hit the limit it just throws the injectors open at 100% duty cycle, since it's the best guess.

So it's possible that this fuel cut is some kind of rebellion against seeing the MAF sensor at its rail for too long, since that's certainly an abnormal condition. But Jason, you never experienced fuel cut problems after installing my FCD, right? So I don't know what to think... it's true that the code in Phil's and Matt's ECUs is different from the code in the boostjunkie Legacy's ECU, so maybe something changed. I don't know. But it's clear that this is something that needs to be addressed.quote]

It is interesting about the MAF sensor. Since the 91s utilize the same MAF sensor as the 92-94s, I would expect the same to be true for my car back in the day. This would also explain why I would also hit 100% duty cycle at anything over 12psi at 4000rpm. Seems as though that was the thrershold for reaching this 200 g/s mark on the MAF.

However, you are correct that I NEVER experienced fuel cut once your FCD was installed. I could boost as high as I wanted (until motor failure, it seemed). Even under extended periods of time, I would realize the boost I wanted without ecu intervention. This obviously dispells the hypothesis that extended maximum MAF readings cause ecu intervention, at least in the 91s.

And we did end up double checking my boost gauge compared to ecu readings. They were almost identical, so the ecu was reading the proper boost until 13psi (or whereever you had the voltage clamped).

[vrg3]

If the MAF is a big detriment, is that the sensor's range or the ECU's? How do people run big numbers with aftermarket ECUs if it's the MAF's limitation?

Most aftermarket ECUs are speed density and don't use a MAF sensor at all.

I think there are a few people who use higher-range MAF sensors.

Actually, the stock black plastic JECS MAF sensor used on 92-94 non-turbo and 90-91 4EAT non-turbo cars has a range about 50% higher then the aluminum Hitachi MAF sensor; it goes up to around 300 grams per second.

This would also explain why I would also hit 100% duty cycle at anything over 12psi at 4000rpm. Seems as though that was the thrershold for reaching this 200 g/s mark on the MAF.

Yup, that's just what Matt and I were thinking. 12 psi was probably right around the threshold.

Even under extended periods of time, I would realize the boost I wanted without ecu intervention. This obviously dispells the hypothesis that extended maximum MAF readings cause ecu intervention, at least in the 91s.

Yeah.

I wonder if we should try swapping the boosjunkie Legacy ECU into Phil's car at some point.

I wonder if it's at all possible that the fuel pump had something to do with it though. If Phil's and/or Matt's stock fuel pumps eventually fail to keep fuel pressure high enough, the ECU might momentarily (too short to see on the scan tool) see a not-rich signal from the oxygen sensor and cut fuel.

Hm.

Phil, since you need to do this anyway, I think you should get a Walbro and install it right away. And let's see where that goes.

[BAC5.2]

That's one of my next Mods. I just need the money to pick my buddies up (He's got one sitting in the garage for $80 bucks for me, brand new).

Anyone want to buy a turbo IAC valve or a roof rack?

Is there any way to re-wire the stock MAF to read higher voltages?

I seafoamed my engine (and dumped the last 10 ounces in a new tank of gas), did a compression test, changed my plugs, air filter (still gotta get a K&N, lol) and oil filter. So right now, the car is running pretty nicely, and now would be the time for me to get a beefy fuel pump and possibly crush the cap of the FPR. I'd like to have an IDC gauge or something though, so I could at least see what's going on with the injectors.

I'd like to get my hands on a 300Z-TT MAF or possibly see about a MKIV Supra MAF (since your brother mentioned the ease of finding them).

OR... I am recalling an older Sport Compact Car about a car that used TWO 300Z-TT MAF's. I'm curious if that would be an option, or more importantly, if it would properly work with two stock Turbo BC MAFs. I'll look through my back mags, and see if I can find the setup. It looked pretty legit, and the car was making some rather large numbers, but I don't remember specifics.

[vrg3]

The only potentially reasonable way to increase the range of the stock MAF sensor would be to make a new housing for it with a larger cross-sectional diameter.

Using multiple sensors in parallel could work but it would be weird to plumb with our single turbo.

From what I understand, a 300ZX MAF sensor (apparently both the turbos and non-turbos in the 90s used the same one, and so did the J30) has a range up to somewhere around 450 grams per second. The part number is 2268030P00 (notice how it starts with the same five digits as our MAF sensor part numbers?). It also has virtually the same shape as our stock sensors. Sounds like it could be a good replacement with the proper supporting electronics.

[georryan]

Either that or buy a computer from Kelley and do away with the maf setup.

Although, if we could easily change out our MAF it sounds like our cars could easily gain a lot of untapped potential.

I"m still wondering if there is some difference between the 91's and the post 91's in the computers themselves. The MAFs are the same, but I wonder if the computer in the 92-94s have some extra lines of code that checks for the MAF maxing out.

Matt and Phil had the problem but Jason didn't.

[BAC5.2]

Meh, plumbing is no real problem. I'm looking through my stack of SCC mags trying to find the dual 300Z MAF car. They used 2 pod filters (one on each MAF) and joined them. A trip to Home Depot and I could figure it out without to much trouble, I'm sure.

What about the MAF from a WRX or STi?

Z32 MAF's are like $100 bucks on E-bay...

http://cgi.ebay.com/ebaymotors/ws/eBayI ... eName=WDVW

[georryan]

Oh, yeah, I'd definately try to run the drag again with a new fuel pump, then run it with boostjunkie's ecu if the fuel pump doesn't help.

[vrg3]

Either that or buy a computer from Kelley and do away with the maf setup.

Well, even with a different MAF sensor you wouldn't be able to just use the stock computer.

I"m still wondering if there is some difference between the 91's and the post 91's in the computers themselves. The MAFs are the same, but I wonder if the computer in the 92-94s have some extra lines of code that checks for the MAF maxing out.

Yep, that's what I'm wondering too. I haven't had a chance to dump the code of Matt's ECU, or the ECU in the boostjunkie Legacy.

I do know for sure that they do have some slight differences in how they handle injector pulse width calculations. For one thing, the 91 ECU doesn't format the pulse width for the Select Monitor, while the later ones do.

Meh, plumbing is no real problem. I'm looking through my stack of SCC mags trying to find the dual 300Z MAF car. They used 2 pod filters (one on each MAF) and joined them.

I thought with a 300ZX TT they could just use one sensor for each bank of cylinders...

A trip to Home Depot and I could figure it out without to much trouble, I'm sure.

Hehe. I'm gonna be a lot more careful about Home Depot car parts now...

What about the MAF from a WRX or STi?

Yup, that's another option. 'Memmer I said that when we were at W&L Subaru? We'd just need to figure out a way to mount it in place; from what I understand it doesn't bolt up the same way as our stock one.

[BAC5.2]

Ahh, that's right, you did mention the GD MAF.

Yea, Home Depot car parts can come back and bite you in the butt. My sewer pipe has been OK, but I think it's because my pipe butts right against the IC inlet. I need to find a 2.5" SAMCO style hose.

The car using twin 300ZX TT MAF's was not a 300ZX TT. It was just a car, that happened to be using two of those MAF's. I haven't found it yet though. I will.

[BAC5.2]

OK, found it. The dual Z32 MAF's were on a Nissan Skyline (R32) in the USCC. It made 661awhp, so apparently it workes pretty well. It also won second in the power delivery/band competition (second to an 842whp Dodge Viper), so it's not just for peak power.

The setup looks fairly easy to fabricate, just a Y-pipe with equal length "legs" with the MAF's samco-style installed. Then two K&N Cylinder filters are on the ends (one for each, duh).

Single turbo conversion, I6, it'd be the same setup style that we would use. But is it worth all that work to get, potentially, 400 g/s? Or could we just do an N/A maf, good for 300? Or two of the N/A maf's? One Z32 MAF? I think we should worry about how to set it up later, we just need to find the best option then work it out.

[NemesisEJ22t]

The only problem i see about using a different MAF would be the fact that the values may be different. Lets say that 200g/s on our stock MAF translates roughly to 4.5volts. Ok, now say you swap a z32 MAF into your car and try to run it without any sort of "translation" equipment. Maybe the z32 MAF has 400g/s set at 4.5volts as well. So you see, your computer would think its only getting 200g/s of air when its really getting twice as much, the lean A/F that would result probably wouldn't be too pretty. Another option i will look into comes straight from the DSM's. They move their MAF sensor to a position past the turbo to essentially make a "blow through" design. I think this is mostly to allow them to vent their bovs to atmosphere, but i will check it out as i know a few who have done this and are recirculating their BOVs. We can solve the problem, it will just take a little experimenting.

[azn2nr]

phill your pass was only 15 seconds but your trap speed are high 13. just solve your fuel issues and youll be flyin

[BAC5.2]

Ben - The post-compressor MAF is mainly used for venting purposes IIRC. I wouldn't run it like that simply to prevent oil from fouling the MAF. The GM 3" MAF that they use to do that has no problems dealing with 400+whp of air. We would need a translator for the Z32 MAF, as we would with ANY non-stock MAF. Even if we used 2 stock MAF's in parallel we'd need some form of translation device.

Jason - If I get my launch issues taken care of, and get 16psi running safely, I should hit the gates in the mid to high 13's and close to 100mph. Then it's just a matter of tuning and cooling, and I might get down a bit further .

My highest trap of the day was 90.93mph, which is dead on with stock WRX 1/4 mile traps, and they run around 14.3-ish.

[vrg3]

Ben - Remember, we're not talking about using the stock ECU with this. That's why I mentioned "supporting electronics." You're absolutely right that a different sensor would have a different transfer function.

I believe the popular MAF sensor for DSMs is a 3" GM one; they place it where you're describing and use a "MAF translator" to convert the signal to one compatible with their stock ECU.