Check the clutch switch.
You are likely to pull the tranny to check the clutch.
Check the fuse box.

 

Unfortunately man it may be the dreaded C-word. If it isn't, I'd say its flywheel issue of some sort, as lightweight flywheels drop RPM quickly too. Maybe you lost some teeth? Here's a pretty good thread on tooners:

http://www.dsmtuners.com/forums/drivetrain-tech/238013-big-problem-push-clutch-car-stall.html

I'm pretty sure the in out play isn't a problem. I'd be more concerned about side-to-side and up-down play. Maybe someone with a bit more expertise can help. Best of luck man, hope it's nothing too serious. Good reason to build a stroker though....

 

... I'd say its flywheel issue of some sort, as lightweight flywheels drop RPM quickly too....

When changing from a stock flywheel to a lighter one, my car would stall out sometimes, but not as bad as the op's.
This was NOT soley the flywheels fault, it was just that its presence made the other issues (slight boost leak, tune, base timing, etc) more pronounced. The car ran fine with these problems while running the factory flywheel and the ecu had time to "catch" the falling rpm. With the other problems fixed the car ran fine with the light flywheel.

 

 

I was about to say that but the picture is worth a thousand post! lol

 

Crankshaft endplay spec; .0020-.0071 in. (according to my source, someone please correct me if this is innacurate).

If it is a crankwalk issue, I would have to assume that the ECU is losing a crankshaft position signal due to a tone wheel moving away from the sensor pickup. I'm not sure what kind of tools you have access to, but a relativley non-invasive way to confirm this would be to put a labscope on the crankshaft position sensor signal output, and see what happens to your square wave as you begin to push in the clutch. Obviously, if the engine dies, you will lose your output regardless (as the engine is not spinning). What you would have to look for is an erratic signal that gets progressivley worse as you depress the clutch (look for rapid spikes/troughs on the output).

 

Just out of curiosity...are you recirculating your BOV?

 

First thing I thought when I read the thread title.

I'd be surprised if this were it, though, seeing as from the sound of it, it'll do it in vacuum, too.

 

My money is one crankwalk... What kinda clutch is in the car, out curiosity...

 

Well, it's definately looking like the dreaded crankwalk. I started working on getting the oil pan off, that'll give me the for sure answer. Either way, what can I do if the car has crankwalk? My current plan is to just replace the bearing, as in another month I'll be driving the 3000GT and can devote more time to fixing the talon.

I know that is just a band-aid and not a solution, but exactly how long will it work for? More into the long run what can I do to cure this crankwalk issue? I'm not too keen on swapping to a 6-bolt, so what are my other options? All I found in searching was the swap but that's a lot of time/money that I don't have atm.

Whatever the case, I do plan to keep the car so I'll fix it one way or another. Thank you all for your help, I'm hoping the replacement will at least make it drivable for a month until I swap cars. Thanks again.

Melbowski

 

I'm not too keen on swapping to a 6-bolt

Ok, :dunno: well you can use an eagle crank in another seven bolt when you rebuild, that should lessen the chance of CWing a 7bolt again.

 

Update on the Talon, finally! Pulled the pan and pulled out the bearings, and oddly enough these look like the stock bearings... The only thing stamped on them is D78 no brand or anything. Either way the thrust bearing is ground to crap on the clutch side, other side looks very minimal wear. This was definately giving me cranwalk, so at least I know what the problem is.

On the side, perhaps I'm being optimistic, but these bearings look like they are the originals. If they are then with 155k perhaps it is long-term wear. When I bought it the guy said it was rebuilt because they did it when they swapped in new rods/pistons (he wanted to do autocross) so he could up the boost, but lost interest and never got around to getting DSMlink so he just used it as a DD. It doesn't burn any oil or have any smoke, so I'd assume at the very least that the VSS have been replaced. That said perhaps he was working from the top-down and therefore didn't replace the crank bearings. If so maybe I'll come out of this alright, it's a long shot, but one way or another I'm not ditching this DSM.

The other end is perhaps I just got raped on the car and he didn't do dick but abuse it (probably more likely), we'll soon see since I'm going to finish replacing the bearings. I'll drive it until it dies again, and if it pukes I'll just do a 6-bolt swap over the summer. I can't really complain since I have two other cars to drive until the Talon is back on its feet.

Like I said, I'm not going to ditch the car, it's a true 98 TSI/AWD and last I knew they're pretty rare for that year so I'd like to hang onto it. I'll nurse this thing back to health and be another proud member of the "My DSM Actually Runs" club :burnout:.

Thanks for all your help so far guys, I really do appreciate it. I'll keep the updates coming as things come up. Thanks again.

Melbowski

 

Alright, this thread has been stagnating for a while, but a new developement has arrisen. That being that I've replaced the bearing and less than 10 miles later I'm still having the exact same issues! WTF!? I know CW is a beast but last I knew it took thousands of miles (about 5k or so from what I've read) to get the symptoms back. That said I'm really starting to lean trowards CW not being the real problem here.

Anywho, this weekend I'm going to replace the crank sensor and pray to the DSM gods. Maybe I'll get lucky and this whole thing could have been a blessing in disguise since the other bearings were obviously worn. We'll see what happens, and as always I'll keep updating this thing, if for nothing else but for some later user to learn from my mistakes, or perhaps from my success. Thanks again to all those who have posted.

Melbowski

 

I really think an engine swap is a good idea because for most engines once it catches the cw bug it seems to be progressively more common, but maybe I'm being pessimistic. Check out jackstransmissions.com they have a warranteed 7 bolt rebuild that garuntees to CW, they probably remove some of the oil squirters to get better oil pressure as that seems like the general idea as to how CW develops, lack of oil on that bearing.

 

Not to thread jack but I’m right there with u man. My car died right after making a loud ticking noise when I was trolling through a parking lot, she turned back on and the noise was present for a few seconds then went away. I drove about 3 miles and she dies out, crank no start. I replaced the crank sensor now she starts and has the constant ticking noise and when slight pressure is applied to my clutch noise gets louder then dies if I press clutch all the way in. I’m tryna be optimistic as well and hope it’s a clutch problem, Ima start tearing down this weekend if I can and see if I can lend some useful info.

 

Glad to see I'm not the only one fighting my car, seems like I've been messing with the talon forever and getting nowhere. As for your GSX if you're hitting the crank sensor, then you definately have excessive end-play. I had 30 thousenths (.030) of play and it never touched the sensor. Check how much play you have and go from there, it's much faster than dropping the tranny to check the clutch. Good luck to you, hopefully we can both have drivable DSMs.

Current status on the Talon is still down and out. End-play is at 5 thousenths (.005) so it's in spec, so that's not my issue, at least not atm. I built a pressure tester and found massive leaks! I fixed all of those, but the problem has not shown much improvement. The car doesn's die anymore when you push the clutch, only if you hold it in and rev it, it's like it can't catch itself.

Meanwhile, I checked my IAC, cleaned it and it moves just fine. FIAV seems fine, the idle lowers as the car gets warmer. TPS was tested and is fine, and properly adjusted. Even with the boost/vacuum leaks fixed if I crank the BISS all the way in it just barely gets to 750 RPM. I just can't figure out what is causing my stalling.

I bought Evoscan to see if I could find anything that way, but it refuses to connect to my ecu, it just gives me FTDI read error - 4, which I can find no info online for what that means or how to fix it. Likewise I tried using MHIScan to not much avail, it will connect, but the numbers just jump all over the place even when the car is not running. I have no idea why it will not log, as it connects and is read just fine with a typical hand-held OBD-II reader, all the live data is read just fine.

I'm really leaning towards getting a 95 eprom and DSMlink so I can at least log. I'm also getting really close to just saying F it, and doing a swap.

Anywho, thank you all for keeping this thread going, I appreciate the ideas, as well as the comments. Any more suggestions would be great, I'll update as the situation progresses. Thanks again,

Melbowski

 

I second Jacks for a good 7 bolt, they have a guarantee on getting rid of excessive crankshaft end play and I believe they leave the oil squirters in, as they are pretty important.

 

Do you Know how to drive stick?? :really:

If it is Cwalk i really, Really Am sorry for your loss- I hope she gets better soon..

 

did you properly adjust the timing with the connector grounded?

did you properly adjust the idle stop switch?

Did you check the throttle cable for enough free play?

 

The idle switch in the TPS is set properly, and the throttle cable has a little play and allows the throttle to close all the way to the SAS. As for timing/base idle, I thought 97+ could not be adjusted without the service tool? I don't know for sure, but if I could ever get EvoScan to work, I'd have access to that. Unfortunately I can't find any info on the FTDI read error - 4 that EvoScan displays, so I don't know how to get it working.

We have an adjustable timing light so I can check the timing to at least see if it's close. Do you know what it's suposed to be at when idling at operating temp? I'll search a bit and test it out, and I'll post up my results; thanks for the idea .

 

Alright, an update on the Talon. We checked the timing with the adjustable timing light, looked like it was correct. So I gues that rules out the 3 options from Bohrn.

Meanwhile, we continued to tear apart the interior wiring that the previous owner has all cobbled up for his gauges/security. We found the wires for the narrow-band he has installed, they were conveniently unplugged. It's tapped right off the stock O2 sensor.

Then we decided to install my AEM wide-band I had bought, hoping it would give us some leads, and it did! It turns out that at idle I'm running hella-rich, over 18:1! It goes high enough that my sensor can't even read it! Meanwhile the narrow-band running off the stock sensor just ocilates between rich/lean every second. WTF is going on here? If we unplug the front O2 the A/F drops down to about 12.5:1 and holds there. Plug it back in and the wide-band rolls back up past 18:1 over about 5 seconds.

Interestingly, when you listen you can hear the engine build more power as the A/F goes up, which it should, but then it just keeps on going and you hear the engine bog down as it goes past 18. So I think running hardcore rich is likely what's causing my stalling eh?

The real question is what can cause that? I'm thinking the front O2 must be bad, because the computer still has control, since it pulls the A/F back to 12.5 with the sensor unplugged. If anyone has had a similar experience and solved it, I'd love a second opinion before I throw down $75 for an O2 sensor. Hell I'm not even sure a narrowband can be run off a stock O2 sensor, I don't know much about them. Either way, thanks for all the help so far guys, I'll keep updating this as new developements emerge, even if no one is reading it...

 

Rofl, it dawns on me that I kept writing rich when I meant lean. :wall: However, I can't edit my post, so I guess that's how it stays. Sorry, I promise I'm not really as stupid as I make myself sound!

Anywho, I just got done pressure testing everything and I can't find anymore leaks, so the O2 is still my prime suspect. We'll see what happens, and I'll keep my retardation to a minimum.

 

That is hella lean... Im talking on the brink of dieing lean... You have a fuel problem my freind. Check the injectors and make sure there not 390s or something dumb. New fuel filter ect... Make sure you have a vacumm line on the fpr... You need to fix that lean issue.

 

well, as far as checking the timing, if its on, it rules out the TIMING. It does not rule out the base idle and the TPS/stop switch adjustment.

A narrowband HAS to be run off a stock sensor, as long as the stock sensor is a narrowband sensor, like our cars are. A wideband needs its own sensor. Some newer cars(some VW's) have widebands stock, for thier engine management.

If the stock O2 sensor is able to correct the a/f while in closed loop, I would suggest that its fine.

 

As stated in post 22, the TPS is set 100% correctly (I fiddled with that damn thing for an hour to get it perfect). If by the stop switch you're referring to the SAS, I'd prefer not to touch that as it seems untouched from the factory, but I will adjust it if I can't find any other problems. As for base idle, from what I've read I need the MUT-II scan tool to adjust it properly; which Evoscan would happily do, if I could get it to work...

Granted I have not ruled out every option, but the remaining base idle I don't believe I can properly adjust without the scan tool. I can turn the BISS by itself and raise/lower my idle, but it has little to no effect on the stalling issue. Meanwhile, in closed loop the stock O2 still pushes beyond 18:1 and never comes down. Until it enters closed loop, the car runs reasonably well at about 12.5:1 but then rolls off past 18:1 as it warms up. Also, thanks for the info on the narrow band, I've always heard they were junk so I've never bought or installed one.

The fuel problem idea I can work with, I've been wanting to change out the fuel filter for a while. Injectors are stock-looking black tops with green-tips; I believe that is the 450's. If not, how can I identify the injectors? The vacuum line is attached to the FPR. I'd like to check my fuel pump, the guy I bought the car from said it was a Walbro 190, and I'd like to double-check that. Either way I'm lean so it's not over-running my FPR, but I'd still like to know for sure.

I know there's a lot of info in this thread, I'm beginning to feel like I'm working on a novel... Please bear with me, hopefully I'll get this thing wrapped up and back on the road soon. I appreciate all the advice so far guys, thanks again.

 

Well, I've had some free time lately to work on the Talon so I'll give an update. I finally got EvoScan to work, and it's fantastic! It pulled an error code that the front O2 is having a heater malfunction, pulled it out and it was caked in crap! Cleaned it out and my lean problem, while still here, is significantly better (no more off-the-scale readings), so a new O2 is coming in soon.

Meanwhile EvoScan shows that my Idle Switch in my TPS isn't getting a signal to the ECU, it just remains at 0 so something is funky there. I know for a fact that it's set right, so I'm guessing I have a short, but I'll test it's position anyway. The kid who had the car before me did some hack-job operations on it, so I'm guessing I'll find that the wire is cut or something rediculous. I'm hoping that this will solve my problem, seeing as the car never knows I let go of the throttle it would make sense that it doesn't attempt to idle. According to EvoScan the IAC never even moves from 40 steps to catch the decel when I let off the gas; further leading me to believe that the car is never aware I've fully released the throttle. Looks like Bohrn was on the right path with the Idle Switch, just in a different area than expected.

I'm going to recheck the switch and then follow the wiring back to the ECU to find the short or cut. I know it's set properly, so something must be interfering with the signal. I'll update this once again after I check the wiring, if only to help another member with the same issue. Thanks for all the help so far, I hope this will do the trick and get the Talon running again.

Melbowski

 

Do not adjust the SAS. Ever! As for the correct setting procedure its the following for the TPS...

2G's critical TPS adjustment is the Idle Position Switch that's part of the TPS.
(2G TPS pins are 1- 5v, 2- TPS output, 3- IPS output, 4-ground)

You place a .0177" feeler guage between the stop screw (fixed SAS) and the throttle pulley to open the butterfly and then adjust the TPS right to the point where the IPS (pins 3 and 4 on the TPS) switches from closed (0 ohms) to open (infinite ohms). Then you verify that the TPS (pin 2 and 4) reads between 0.4 and 1V while the feeler guage is still in place. Anything in that range is ok as long as the IPS is adjusted correctly. It it's not correct you need to check that pin 1 is about 5v and pin 4 is grounded. If they are not correct then the TPS is bad.

There is no way to adjust the TPS voltage independantly of the IPS on a 2G. Trying to correctly set a 2G TPS by voltage will only work by dumb luck.

This is very important for the 2g's The 2G ECU looks for the closed position as its a different fuel map/table for idle. The open position is all the rest of the fuel tables.

I'm glad you are keeping the 98. I too have a 98 and I actually looked all over the US just to find the specific 98 model year. (Had it shipped to me) It is a rare year. Only 4308 were made. That number is the combined made for Base, ESi, TSi and the TSi AWD. So its probably correct to assume that maybe 1500-2000 of them were the top model AWD. Maybe even less. Very rare indeed. Good luck.

 

No worries my friend, I will never touch the SAS unless I know 100% that it has been touched in the past (and the white paint from the factory shows that it has never been touched, so it's safe). I've tested the TPS thoroughly and it's in perfect working order, I have it set to activate the Idle Switch at .6mm from the SAS, as per my service manual suggests (took me like an hour to get it just right).

I believe the problem lies in the connection to the ECU. The previous owner hacked up everything under the dash to install his security system and it wouldn't surprise me at all if the wire got cut. I looked through some diagrams and it appears the Idle Switch runs to pin 87 on the ECU, can anyone confirm this? I'd like to trace the wire to check if it has been cut.

On a side note, Tarantula I just read your thread last night on making the rear TALON letters glow, it looks fantastic! I look forward to seeing the finished product back on your car; if it turns out as good as it seems like it will, I'd like to do it to my car in the future.

Also, I too searched for a 98, these things are elusive! It's sad that we don't really know for sure how many TSi/AWD models were made. I know for the 3/S platform they made an algorithm that ran through the CAPS program to discern all the VIN's, I'd like to know if they could set it up to pull all the DSM VIN's just to see how many of each trim were made. As for my car, it unfortunately came from Wisconsin and rust is already very present and likely unstoppable. If it's fixable, I'd like to restore it, if not, once it's completely rusted out I'll likely swap the sub-frame into a 2gb Spyder and convert it to a Talon (using your guide none-the-less) so my car will be reborn as a convertible.

Back on topic, thanks for the advice, I'll recheck the TPS to be 100% sure that it's correct before I begin to trace all my wires back to the ECU. I'll update this thread again after I fix the connection problem. Thanks again.

Melbowski

 

Yeah I have had the same problem with my Talon ESI. Obviously it couldnt be crankwalk for me lol. BUt it would happen at random times but especially when cold?! It hasnt happened for a while now and i guess it doesnt matter now because i sold it lol