HELP! Car surges/falls on its face around 4000- 5000 RPM

[Maxboost1010]

Car is a 1998 GSX. 2.0L Turbo All wheel drive.
Hey guys I'm a little new to DSM but I'll do my best to explain my problem and hopefully someone can help. When you go Wide open throttle and get in the 4000 RPM range the car feels like its hitting a brick wall then fine then hitting a brick wall over and over again and can't seem to even get past 5000 RPM. I made a, homemade boost leek tester and found leaks at the blow off valve, which I JB welded, and found a little bit of air escaping by the throttle body, is this normal or not? I'm going the replace the gasket their this weekend anyways [I think their is only one gasket, right]. Other then that today I replaced the spark plugs with new NGK BPR7es and new plug wires with Accel 300+ Turbosport. That haven't helped either. ONLY once in a while I get a check engine light coming on and then after driving it for a while it goes away, the code is for is P0170 Fuel Trim Bank 1. Also I have noticed when I'm on Wide open throttle it blows black smoke out the tail pipe, so I'm running rich as hell I guess. I'm running out of ideas, any and all help would be great. By the way the car has 138K miles on it do you think the cat could be causing this problem, If I'm not hard on the car it drives fine. Also I did a compression test on it today, and I got cyl 1- 150psi, cyl 2- 160 psi, cyl 3- 155 psi, cyl 4- 152 psi. My firing order is set like this standing in front of the engine looking at the engine. Pistons 4, 3, 2, 1 then the timing belt. Looking at the coil pack the same way [from passenger side to driver side]
4, 1, 2, 3 I'm assuming the firing order is correct.

 

[hpexperts]

hmmmm sounds like its retarding the timing. if you are getting lots of smoke with a cat i would diconect the down pipe and take it for a spin to see if it stops. how much are you boosting? how did this all start?

 

[hpexperts]

when you hit this brick wall does if feel like fuel cut?

 

[4Motion4g63]

Im gettin Something Like this Also, Just Installed a 16g and still have stock fuel system, It Pulled Like a Raped Ape for the first week of the NEW turbo and FMIC install but after that its been Pulling timming and my FUEL trims are MAXED out.....Sometimes It just Pulls hard again. I also Have a Brand NEW Knock sensor....

 

[Maxboost1010]

Well I just bought the car like 3 weeks ago, and it's been happening since.. it's always been happening. I'm kind of new at this dsm, how do i check the timing being retarded, do i just use a timing light? or is there a computer program or some sort of scanner I can use? Do most people just eliminate the cat and put a test pipe in it's place permanently? As far as boost, the car is stock so it looks like a standered 14b turbo on there with no boost gauge. I'm trying to fix these issues before I invest in gauges and all that stuff. What does a stock turbo boost at, is it 10psi? As far as the fuel cut goes, I'm not sure, I've never had a car do this before. Can I check by taping a fuel gage to the windshield and seeing what happens? What else can I do?

 

[hpexperts]

no you need a data loger to see what the car is doing like fuel trim and knock, timing etc... what is your set up?

 

[Maxboost1010]

Yep I agree, I bought Evoscan yesterday and the cable for it should come in in 5 to 10 days, I'm going to hold off till then and I'll post up some results of the datalogger when I get in a few days. Any of you use evoscan as your datalogger?

 

[98GSXtreme07]

Well you've already PM'd me but you know i use it. If there are any questions you have about the logger than just go here: http://www.dsmtuners.com/forums/newbie-forum/271955-evoscan-does-anyone-use.html

 

[Maxboost1010]

Originally the car is completely stock, The previous owner had this same problem and replaced it with a used turbo, so I'm not sure which one's on there right now. I bought a bunch of new parts, which are the upper intercooler piping from dejon (not the s-pipe yet though), and bought a greddy type-s blow off valve that fits the pipe, I put rtv on the gasket before I put the BOV on. I also got rid of the stock air box and replaced it with the K&N fuel injection performance kit. I also got a evo-scan data logger, I just logged it today for the first time ever, and don't fully understand logging to the extent of other people on here, so you guys please help me out, and tell me what i'm seeing. Open the log in excel, and I start getting on it around line number 224, and start letting off around 404-405. That was on the highway, and it was on the highway alittle before 224 and after 405, but that was the main time I was heavy on the gas, (the most rpm I made it up to according to the logger was around 4600 rpm). It shows rpm, and it was letting us read the speed while we were driving on the logger, but on the log, it doesn't show the speed for some reason, and I think the boost on the log is wrong. Also I bought a glowshift boost/vacuum, oil pressure, and exhaust gas temperature (egt) gauges, and will try to put them on this weekend. I also baught a SBR MBC that I will put on once I have a boost guage. Other then that the car is stock. Let me know what you guys think.
The only way I could figure out how to upload the log was through rapidshare, it's completly free to view and upload/download, if you guys have an easier way, let me know. All you have to do is go to RapidShare: 1-Click Webhosting
Then click on free, then just wait the amount of time, (normally around .5 seconds), I don't think it matters which download from you use, type in the characters it says on the bottom, and click on download via..., then the pop up window for open or save will come up, and you can use excel for it, remember to focus between 224 to 406. Sorry for all the steps, and like I said if theres's an easier way, please let me know.
Thanks

 

[Maxboost1010]

Need Help With This Log! What Do You Guys Think?

 

[Maxboost1010]

Anyone?

 

[Maxboost1010]

Ok I tried hooking the vacuum source straight from the niple on the turbo housing to the wastegate to bypass the actuator, my stock boost sits between 13-15 psi and when I hit Wide open throttle it throws me back and forth in my seat (will not make it past 4600RPM no matter what I do) I agree its probably some kind of fuel cut out issue, I do not have any type of d-link or stand alone system and didn't think I would need it on a stock car. One thing to keep in mind is I'm not the first owner of this car and it had the same issue since I have bought it, do you guys think the previous owner could have been messing with something that would have caused this issue to happen, and what component would he have screwed up? I've done boost leaks, fixed what I could find, some piping and couplers and still no luck, I got a SBR MBC and have it set to the lowest setting and still no change ( I know a MBC won't lower boost below stock levels so that wouldn't have an effect anyways) basically when I'm getting thrown back and forth in the seat the boost guage is fluctuating back and forth between 15 and 13 psi extremely fast. I have even tried having the wastegate held open with a vice grip to get the boost down to about 10 psi and still no luck, any ideas at all guys to where I should look at next?

 

[Maxboost1010]

Ok I have been reading anything and everything I can get my hands on, I came across http://www.dsmtuners.com/forums/newbie-forum/225897-sputtering-hesitating-under-boost.html which sounds a lot like my problem, and for him it turned out to be Barometric sensor so my next question is on a 2g is the barometric sensor part of the maf or somewhere else and how do I test it, I can't seem to find any post on that location.

 

[Maxboost1010]

Ok I did another boost leak test and would like to get your guys opinions, I found a leak at the throttle body, I replaced the rear gasket on the the throttle body (the one closest to the intake manifold), still have the leak (see bubble in the pic) so I'm thinking its the throttle body shaft seal, can anyone confirm this? Will that leak cause the problem I'm having? Next question, I bought from a used car dealer, so it looks like who ever owned it before has a bigger turbo on it then the one that's on it now, and he had wires running for the gauges under the dash but those were missing, so I don't know if their is a bigger fuel pump or bigger injectors in the car right now with no AFC (maybe he had one and took it out, don't know) could this problem happen if the due to having a bigger fuel pump and bigger injectors, or am I thinking back-words and the opposite would be true?
(also the car is running pig rich at WOT)

 

[bmoha2]

and get an afpr into that car It doesn't sound like it has one.

 

[Maxboost1010]

and get an afpr into that car It doesn't sound like it has one.

I'll defenatly look into getting one, which one would you suggest? How can I find out if it ever had one and it was taken out? I'll just replace the pvc tommorow do you think that could cause this type of problem? with an aftermaket fuel pressure regulator (afpr) can I adjust for how rich it is running or do I still need to get an aftermarket fuel controler (AFC)?

 

[SNE2040]

i have the same issue with my small 16g. my car starts boost creeping once it hits over 4500 RPMs and i also just started getting P0170 (which can be a fouled out O2 sensor in the O2 housing). My car boosts ~13 and then starts flying to ~20-21 PSI and then it hits fuel cut. I'm also running stock fuel system with the bigger turbo and 3" tbe and intake

 

[bmoha2]

Look into getting your wastegate passage ported. There are also ideas out there that if you went to an atmospheric dump O2 sensor housing it might alleviate some of that problem.
Maxboost; I have the aeromotive setup and it seems to work well. It's kind of expensive. There are other options that are less expensive. Mine is setup with SS braided lines. You can get other setups with a less expensive afpr and rubber lines sorry for the dirty car

 

[Maxboost1010]

Ok I'm going to order the throttle body shaft seal on Monday, I'm hoping fixing that boost leak will let my car run right, guys how can I compensate for an over rich mixture, whats the easiest and cheapest way to lean out the car without causing damage to the car? and yes I'm not going to try that till I first fix all the boost leaks.

 

[Maxboost1010]

I finally did the throttle body seals, now I have absolutely no boost leaks. I checked it at 15psi. But the car still won't make it above 4600 RPM at Wide Open Throttle and still blows massive amounts of black smoke through the exhaust (ONLY AT WOT). What do you guys think I should look at next? Just me thinking out loud, I'm not sure what the person who owned it before me had, if they had a bigger fuel pump and injectors installed and they might have taken out the AFC or something would that cause the problem I'm having, and if so will me just getting an aftermarket fuel pressure regulator be able to compensate for the high flow of the pump and injectors?

 

[kenamond]

Where to start...

Can you contact the previous owner? If so, ask them for every mod they did to the car and ask them if it ever ran correctly for them (maybe they bought it, had the same problem as you, then sold it). You're getting too much fuel, and it might be due to bigger pump and/or injectors but no AFC or other fuel controller. I don't know how unique the injector colors are, but you might be able to search for pictures of the stock injectors and see if there are differences on yours. The fuel pump is harder, because it's mounted inside the gas tank. So if you can find out all upgrades from the previous owner, that'd help us zero in on your problem.

So you're getting too much fuel at WOT, that much seems obvious. It also sounds like you're hitting fuel cut. In case you didn't know, fuel cut is caused by too much airflow. The ECU detects that the motor is flowing too much for the stock fuel system and cuts fuel to the injectors completely until the airflow drops back down.

Also, that diagnostic trouble code for fuel trims means you're too rich when cruising. The ECU has a fuel map that stores how much fuel to add for every combination of airflow and RPM. When the motor is cruising, the ECU goes into closed-loop and learn-modes. It starts out with the fuel map values. Then it uses the front O2 sensor signal which will be 0.5v at an AFR of 14.7 (stoich). If the O2 sensor reads on the rich side (more than 0.5v) it removes some fuel. If it reads lean, it adds a bit of fuel. It's "happy" when the O2 signal oscillates around 0.5v. If it has to consistently remove or add fuel, it stores this as a correction called a fuel trim. If conditions are right, the ECU memorizes the correction (that's learn mode). The immediate corrections are stored in a short term fuel trim(STFT), and the learned trims are stored in the long term fuel trims (LTFT-Lo and LTFT-Mid). The Lo trim is used for low airflow and the LTFT-Mid are used for medium airflow. When you go WOT, the ECU relies strictly on the fuel map; it's not listening to the O2 sensor anymore. That's open-loop mode. Back to closed-loop. When that trim gets bigger than (not positive of the number, but it's close) 12.5%, it throws that error code. The ECU will not let the trims go past that value. That means that at times you're running 12.5% rich while cruising (while in closed loop or learn modes) and the ECU is having to remove a lot of fuel from what the fuel map says the motor should need.

So...you're running rich at cruise, and you're hitting fuel cut. This sounds like a boost leak to me. The MAS measures the airflow (that's why it's called the mass airflow sensor). The ECU uses that and the RPM to figure out fuel. To hit fuel cut on stock injectors, you're either overboosting (which will cause you to flow more air than the ECU will allow) or you're leaking air that the MAS measured, and that leak is happening between the MAS and the motor...so the ECU injects the fuel based on the MAS, that's too much fuel, you run rich. If you leak enough, you hit fuel cut. And that's what's happening to you. *aaand* you're running rich during cruise when you're not boosting hard, so I don't think it's overboosting. Since you hit fuel cut, I don't think it's bigger injectors or fuel pump; those have *nothing* to do with what the MAS reads, and airflow determines fuel cut.

But...you say you did a boost leak test (BLT). So I have to ask for details on the BLT results. You pressurized to 15psi and then what...how fast did it leak back down? Getting to 15psi isn't the goal; holding 15psi is the goal. Leaking 1psi every few seconds would be good enough. Did you test at the compressor inlet or the TB elbow? Test at the compressor inlet, because that tests all components that see boost.

Since you have EvoScan, log a 3rd gear pull from 2k to redline (that'll be 95mph at redline, mind you). Log rpm, airflow, front O2v (maybe called bank 1 or something), timing advance. Cut/paste it into a text file or export it in CSV format (comma separated format) and make it pretty in plain text. If you get it pretty, plop it into a post. That's the usual way to log WOT (open-loop) conditions.

Also, monitor your STFT, LTFT-Lo, and LTFT-Mid trims. It might only show you one LTFT variable as an option in EvoScan (not familiar with that logger). The STFT will change constantly but should be small. The LTFTs will probably not change unless you switch between the two. I can rev my motor in neutral and watch the logged LTFT switch from the Lo to Mid. Post what the values are.

While cruising or idling, monitor your front O2 volts. At cruise/idle, the ECU is in closed-loop or learn mode, so it should be trying to get the O2v to vary around 0.5v. If it doesn't oscillate, it might be a bad O2 sensor, but that shouldn't affect WOT rich/fuel cut. But it's something easy to check.

Finally, read all stored codes. Make sure there aren't any others besides the fuel trim code. If there are others, post them.

So that's probably enough for now

 

[Maxboost1010]

Thanks man, that's alot to consider, I'll try to take a pic of the injectors and over the weekend and I'll do a log like you suggested over the weekend, until then here is a copy of the first log I ever made with the car, so far I'm not seeing anything abnormal with it but again, I'm very new to making logs and might be over looking something. Unfortunately its not possible to get a hold of the previous owner since I bought the car from a dealer and they didn't seem to be the best authority on DSM's, I suppose it is very possible that who ever traded it in could have had stuff done too it since they had all sorts of wires running under the driver under dash panel, and when I got it it didn't have any gages or MBC or anything so they may have taken off anything they could sell off.

 

[kenamond]

I wish I had evoscan.

I'm not sure if the Airflow column in the 2nd image is in Hz or some other units. The MAS reports airflow as a frequency...Air sort of buffets around a small bar hanging off of the roof of the inside of the MAS, swirling one way then the next making whirlpools of air behind the bar. As the speed of the air goes up, the frequency of the swirls goes up. There is a probe downwind of the bar that senses each alternating swirl. That probe sends a Hz signal to the ECU. The ECU knows how to convert that frequency into airspeed. From that, barometric pressure and intake air temperature (both also read in the MAS), the ECU can compute how many pounds of air per minute are flowing through the MAS at any moment (that's airflow). I don't have a sense for the Hz values. My logger reports airflow in lb/min. So you might want to search and find out what Hz value relates to fuel cut. It looks like you hit your problem when the airflow value hits 1162.5, but I don't know if that is a Hz number or what (might be grams of air per minute instead of lb/min, or it might be Hz, or it might be something else). If you find that fuel cut happens at 1150 Hz, that'd make me believe you're hitting fuel cut. Normal airflow values in lb/min will be 10-30 for a stock or moderately upgraded turbo.

I don't know what the trim values mean on your logger. One of them is -4ish, and that looks like a normal fuel trim value (you want it within +/- 5 or so). It would mean that you're running rich a bit, and the ECU has to pull 4% fuel from what the fuel map says. The other two look different than I'm used to.

O2v look normal. It's almost impossible to deduce AFR from narrowband O2v, but usually, you want it to be in the high 0.9s at WOT...which is what your log shows. So you're not leaning out on a pull (if it drops into the 0.8s or lower at WOT, you're probably leaning out).

I see no knock to worry about. I think the ECU ignores knock counts of 2 or less, and yours are all less than 2. But I see 0 degrees of timing advance in row 102 which is weird; your timing should be constantly increasing from when boost first hits all the way to redline. Maybe that's just a weirdness in evoscan. I don't see any knock that would do that.

Intake temps and coolant temps look fine. TPS shows you had the throttle mashed for the pull, which is what you want.

You might want to go find that "evoscan, does anyone use it" thread (think it's on the Newbie Forum) and ask someone with evoscan if they can post a log and see what's massively different between yours and theres or ask them if they know more about what units the values are in and how to convert them to more common values.

So much information...but I don't know how to interpret most of it without an expalanation of what each column means.