Resolved New project, diagnosing brick wall at 4k

[daughters-car]

So my daughter totaled her 97 Spyder GST A/T a couple months ago after owning/driving it for over 5 years. It was bone stock and she loved everything about the car except the front wheel drive in bad winter weather. So while looking for an awd car for her I came across an awd swapped 97 Spyder for what I thought was a fair price and we picked it up (and we still have her old wrecked car for parts). Unfortunately the person we got it from had not owned the car for very long and has absolutely no knowledge about the car.

So, our new project is a 97 Spyder GST with an awd swap, manual transmission, and significant other modifications. It’s a running driving car – sort of. I’ve been digging through the car trying to determine what all is there, what is modified, etc and researching for a couple of weeks now – seems we have purchased a real mess since we don’t know the history. The biggest issue I’m currently trying to tackle (and failing at) is a brick wall at 4,000 rpm - seems like fuel cut. It appears the car has a 14b with one of those aftermarket 7cm exhaust housings, but I am only trying to run straight waste gate boost – hitting around 10-12psi right now. I’ve done repeated boost leak testing, fixed several small leaks, but made no difference. I’ve also tried a different known good MAF sensor with no change. The car also has hard intake and boost pipes, aftermarket O2 housing, 3” turbo back stainless exhaust, fuel pump re-wire, and a stock 95 eprom ecu (not link). The spark plug wires have been swapped to run the 95 ecu and it seems the injector wires have been swapped around at the ecu as well, but I’m not sure it is correct. The car seems to idle and run fine as long as you don’t get into it too hard. From the research I’ve done, it seems I should not be hitting fuel cut with a 14b.

When we first got it, it had check engine light codes for random misfire and evap solenoid. The random misfire went away after clearing the codes and I fixed the evap solenoid, but now I randomly get a P0202 cylinder 2 injector circuit even though there is no noticeable misfire. I’ve ohmed each injector and the wire harness from ecu to injector plugs and found no issues. The injector plugs are new, as is the injector resistor pack, power transistor, ignition coils, plug wires and plugs – all done before we got it. Seems the previous owner was possibly chasing this same issue. The car smells like it is running rich, but the wideband gauge is not currently working to tell for sure. There do not seem to be any piggyback tuning devices currently on the car.

So, we have a 2g Eclipse with a 14b turbo hitting a brick wall at about 4,000 rpm under heavy throttle and about 10-12psi boost. It feels like it’s going to break the drivetrain it’s so violent. It will hit 4k and just continue violently jerking until you back out of the throttle. If you very lightly accelerate and keep the boost down, it will go past 4k. I also have a 1991 3000gt VR4 that I log with evoscan, so I’m going to try to get evoscan to work with this Eclipse to see what I can determine from there – other than that I’m running out of ideas to figure this thing out.

Do you guys have any additional suggestions on diagnosing this brick wall?

 

[Red94Eclipse]

have you check the fuel filter? bad/wrong gas?

 

[Turbo2g]

Ive had this happen from a bad MAF before, Do you have the ability to read live data while driving?

 

[lasershark]

Could check the fuel pressure regulator and vacuum lines, make sure they are routed correctly. I have put them on backwards before and had similar symptoms, so it's worth a check. Do you still have the fpr solenoid or has it been deleted?

 

[daughters-car]

Thanks for the suggestions, I hadn't thought about the fuel being wrong - I hadn't put any gas in it since buying it. It was about empty, so I filled it up with fresh 93 octane. Fuel filter looks to have been recently changed. I know the MAF is good because I swapped it with a different car with a known good one. It has had all the vacuum stuff and egr deleted, including the fpr solenoid, so it's not that.

I noticed it looked like the O2 housing might be leaking exhaust fumes and thought the O2 sensor might be picking that up and causing trouble. I pulled the O2 housing to seal it up and found that the wastegate was adjusted wrong and was hanging wide open. Got that adjusted to have a little pre-load on it and got the O2 housing re-installed and all sealed up. Took it for a test drive and still had the same issue at 4k, but it does spool up the turbo a lot better now!

It also threw the P0202 injector #2 circuit malfunction code again. I re-checked the injectors while hot and they all checked out good. I was going to start swapping parts from the wrecked 97 GST when I realized something was different. The 97 has the CAS on the transmission end of the engine, but this car does not - it's on the timing belt end under the cam sprocket. I spent a bunch more time researching and have come to the conclusion this car has the 95-96 CAS setup along with the 95 ECU. I looked into the injector wiring some more and I'm pretty sure injectors 2 and 4 were backward. Pulled the pins at the ECU and swapped them and now the P0202 code has not come back, but the car still runs the same and hits a brick wall at 4,000 rpm!

Since I got sidetracked trying all that stuff, I haven't tried to get evoscan working on this car yet so I'm going to try that next and see if we can get some live data.

 

[pauleyman]

Have you pulled the ecu to have a look?

 

[gsx tuningart]

You have a fuel pressure gauge?
maybe your fuel pump is not good .
i had same problem some years ago .

 

[daughters-car]

I did open up the ecu and it looks ok to me.

I have a fuel pressure gauge, but can't connect it to this car since there is no fuel pressure test port.

I just spent the last 3 or 4 hours trying to get evoscan to log with this car through my tactrix open port 2.0 without any luck. I'm frustrated and done for the day! I kept getting errors about things not being installed correctly, missing drivers and dll's. Finally worked through all of that and now after evoscan connects it immediately stops with an error "J2534 Read Error - 2" and can't get past that. I've uninstalled, re-installed, tried different versions, I don't know what else to try. My handheld OBD2 scanner reads the codes just fine and shows limited live data, but not enough to figure out what's going on with this thing. I think I'm going to have to try ordering a different cable to connect to the car without using the tactrix open port.

 

[lasershark]

I just spent the last 3 or 4 hours trying to get evoscan to log with this car through my tactrix open port 2.0 without any luck

I believe the tactrix only works with the blackbox ecu's with the dsm's. Never messed with it personally...

 

[PlanZero]

You need a black box ecu for evoscan and the extra wire since you're putting a.black box in a pre 98. It will throw emissions CELs as well. Are you positive you don't have a loose IC clamp somewhere? I had a boost leak so big at about 3k rpms (thought it was misfire) I didn't even see it with a BLT. I'm curious as to what else was done besides a 95-96 head and ecu swap by the PO. Almost no one continues to use the 2ga CAS, especially in a swap. Replace that thing with a 1g blacktop CAS, make an adaptor harness from old pigtails and use the 97 ecu for proper firing order (for future reference )

 

[daughters-car]

Thanks for the help guys,

I know the blackbox 98-99 ecu's are needed to use ecuflash and flash tuning on the dsm's, but I thought evoscan worked for logging on any year. I use it on my 91 3000gt with no problems, but I don't use the tactrix for that because it doesn't have an obd2 port. I have the tactrix because I'm planning to switch my 3000gt to a flashable ecu and use ecuflash on it - but haven't had the time to get to it yet. This is the first time I've tried using the tactrix. I'm starting to think about ordering a blackbox ecu so I can go that route, and raise the stock fuel cut limit, then sell the 95 eprom ecu to someone that wants to run link.

I spent the last couple of days cleaning up some messy questionable wiring, verifying the wiring between the ecu and maf is good, and found the throttle position sensor was misadjusted - wasn't tripping the idle switch. Fixed that stuff, but still no improvement.

I found what appears to be a 1g cas and adaptor harness in the boxes of parts that came with this car. I also have the 97 cas from the wrecked gst. What would be the best option to go with here - the 1g cas? I'm thinking about swapping the cas and putting in the 97 gst ecu to see what happens for now. I think read that there are sometime issues with running the 1g cas with the 97-up ecu's?

If the 14b that is on this car has had some kind of custom internal upgrades like a 16g or 20g wheel, would it be possible that I really am flowing enough air at 4k and 12psi to hit stock fuel cut? There was a disassemble small 16g in the boxes of parts along with several loose compressor wheels and other turbos and parts, so it wouldn't surprise me one bit if the 14b is not stock. Any recommendations on how I would tell? I've really got to get some kind of logging solution working on this thing so I can see the actual airflow numbers that the ecu is seeing.

 

[daughters-car]

I was able to get the tactrix and evoscan to work in the wrecked 97 GST, so I think that tells me the software install and drivers are good and the logging issue in the new car is related to the 95 ecu. I am able to use my generic OBD2 code reader to view some live data, so I tried that out last night and I think I can confirm this brick wall is NOT fuel cut. My code reader shows MAF readings in LB/M and the brick wall occurs when that reading hits around 10-11 or so (hard to watch while driving). I can also see calculated load and that is showing around 81% at that same time. I don’t think those numbers are anywhere near high enough to be hitting the stock fuel cut limits from what I have read?

Moving on from there, I did a warm compression test and all 4 checked out good. I also got a fuel pressure test port added and checked fuel pressure at idle. I got 57 psi base pressure and 49 psi with the vacuum line connected. I didn’t check for pressure rise on boost. From those numbers, I’m guessing this car must have an upgraded fuel pump. That probably also explains why it smells so rich at idle, overrunning the stock fpr. Would it be possible that this thing flows so much fuel under boost (guessing 1:1 rise would put fuel pressure at 69psi at 12psi boost) that it’s going outside the ecu’s ability to adjust and actually flooding the engine out at 4k and 12psi boost and that’s my brick wall?

The more I dig into this car, the bigger of a disaster I realize it is. I decided to pull the intake and look at the compressor side of this 14b to see if I could tell if it is stock. Looks like it is, but it also has a bad compressor wheel! One of the blades is missing a pretty significant chunk and the others have obviously made contact with a housing before – but not the one that’s there. I think they must have done a home brew rebuild because there is very minimal shaft play, they must have re-used the damaged wheel for some reason(being cheap).

Now I’m starting to lean towards pulling the stock engine and parts from the wrecked 97 GST and take this new car back to stock condition. Problem is the GST has 240,000 miles on it! It used minimal oil between changes and doesn’t have any significant oil leaks since we kept up on maintenance on it. I’m going to bet the T25 with 240,000 miles on it is going to be pretty tired though. This new car needs cam seals and valve cover gasket for sure and I have no idea when the timing belt was last done. The engine has unknown miles (supposedly lower than the odometer reading as the engine has been replaced), and has obviously been beat on pretty hard. It’s hard for me to not use all these performance upgrades when they are already there though! The end goal is simply a reliable daily driver for my 21 year old daughter though – she’s not too picky except the big loud obnoxious muffler has to go.

Any more suggestions on diagnosing this brick wall before I commit to ripping this mess apart and going back to stock?

 

[daughters-car]

Not making much progress on this thing. I tried testing the wastegate with a bicycle air pump to see what pressure it opens at. It doesn't seem to be opening until around 15-17 psi, but the gauge on the bike pump is hard to read at low pressures like that. I tested a spare wastegate from the spare parts pile and it seemed to open at more like 10 psi. I started to swap them and realized the spare was cracked and about to come apart - so that plan was a fail. While at it, I tested the boost gauge with the bike pump and it seems to be reading accurate. There was also a used afpr in the parts pile, so I tried installing that. But it turned out to be junk, wouldn't hold steady pressure and leaked fuel - so that was a fail as well.

The stock turbo off the wrecked 240,000 mile car is worn out, so that option is out. It's looking like the small 16g from the parts pile may be the best option to put on this car. Just about all of the supporting mods are there. I'll have to send it out for balancing, but other than that it looks usable. I've never had to get one balanced before, what do they do just mark the orientation of everything so you can put it back together the same way as they balanced it?

 

[Vegas Smith]

Stock 2g dsm fuel pressure is around 43 psi. You have stated this car's base pressure is around 57......I think your problem could be there

 

[daughters-car]

Hmm, really? I can understand that fpr overrun would make it run rich at idle, but wouldn't expect it to cause the trouble I'm having at 4k. Since it's overrunning the stock fpr so bad at idle, I gather the pump that's in this car is likely the Walbro 255 which is way overkill. Sounds like I should order a Walbro 190 and swap the pump out instead of adding an afpr - would be cheaper too.

 

[Vegas Smith]

Well, at the very least it's a good place to start and something that should absolutely be fixed asap. Does it not have an adjustable fuel pressure regulator? You can find them brand new for like $150 or on classifieds all the time for even less.

 

[Rekadae]

Yeah, would definitely suggest going for a afpr, because you'll really be kicking yourself if the smaller pump doesn't fix your issues, and in addition even if the 190 does fix the flow rate you'll still end up wanting one later down the road when you're chasing down something else, better to shell out a little more for the sake of usability.

 

[daughters-car]

Well, I'm not making much progress on this thing even though I've been working on it every dang day. This car hates me or something, because every simple thing turns into a major ordeal.

First I decided to get the dead wideband working - figured it would help me diagnose this issue. The car has an innovate DB series gauge in the pillar pod, but the LC1 controller was in the spare parts pile and there was no sensor. Ordered a new sensor and while installing it the bung ripped out of the exhaust trying to remove the plug - got to be about the worst welding job I've ever seen. So off comes the exhaust to fix that mess. While I have that off, I notice a loose bolt sticking out of the t-case. Tried to tighten it up but the threads in the transmission were stripped! Got that fixed along with a bunch of welding on the exhaust, got it all re-installed and moved on to connecting the LC1 to the gauge. Wiring disaster! Spent countless hours re-wiring a bunch of stuff to make it right. Why in the world would someone tap into a 20 gauge or thinner cruise control switch wire to hotwire the radiator fans to come on with the key switch!!!??? That was a melted mess! Way too much amperage pulling from that tiny little wire. Still don't have the fan wiring corrected yet.

I also decided to install the 1g CAS so I can switch to the 97+ ecu without having to swap a bunch of wiring around. I was going to install the 2gb cas from the wrecked car, but the cam in this car isn't tapped for it. Instead of swapping the cam, I figured I'd just use the 1g CAS for now. I'm going to make the jumper harness differently so that I only use the cam signal from the 1g cas and still use the crank signal from the cps. While I'm at it, I cleaned up the valve cover and painted it. Went to re-install it last night and found out the valve cover gasket set I got is messed up! One of the 4 spark plug hole gaskets is the wrong one - check out the picture (one of these is not like the others). So, I had to reorder a new valve cover gasket set and I'm waiting for that to show up before I can finish putting everything back together for some more testing.

 

[daughters-car]

This is funny... I was looking at the Fel-Pro box for my valve cover gaskets. I didn't know you could get a Jeep with the 4G63T engine. I'm going to have to try to find one of those!

I'm guessing they must consider the Eagle Talon a Jeep, too funny!

 

[daughters-car]

Finally got it running right! I guess it was the ecu? I swapped to a 1g CAS and the 97 ecu from the wrecked car and now it runs great. The wideband is showing like 14.4-14.9 at idle, so I guess it's doing OK without the afpr after all. Still not sure if I want to get one or not. This car is going to be my 21 year old daughters daily driver. It needs to be reliable above all else. I see an afpr as another potential failure point. She will be graduating college in 7 months and who knows where she might have to move after college to get a job. I know she won't be around home were her dad can be her free mechanic like I have been for her whole life. She might even end up somewhere with emissions testing and I'm not sure if this car would pass with all the modifications - not something I've ever had to worry about here. She knows how to check her oil and change a tire and that's about it. She doesn't even really know how to drive a stick and this car is a 5 speed. I taught her when she was first learning to drive, but she's had an automatic ever since.

So, back to the ecu, are there any places that buy the eprom ecu's as cores for repair and modification? Since this one is suspect, I'm sure not going to be able to sell it as is. I sure hate to let an eprom ecu go to waste knowing they are getting more rare all the time and are still somewhat sought after.