Long story, so bear with me. This is actually a continuation of a few messages in another thread, but I don't want to hijack that thread.

For reference, see this thread in the Archives.

Let me recap the problem:

Over a year ago, I blew up my engine (my car: 86.5 V6 GT Auto). When I had the garage install and rebuilt another engine, I asked them to also install a high performace camshaft (because they said that would give me a quite noticeable performance gain). The garage chose the Edelbrock 3790. After putting 1400 miles on the engine, it was time to put it to the test. Much to my disappointment, the new engine was not faster than the old engine (acceleration wise), while the old engine had no performance mods at all.

Both engines did 0-60 in 9.4s (avg).

For the new engine, they had set my timing to 4°. After some convincing done by the members here, I had the garage first set the timing to 6°, then 8° and now it's at 11° (and has been so for the past 6 months or so). Strange thing is that this made no difference whatsoever (it doesn't even idle differently - it's still very stable, purrs like a kitten).

During the past 6 months, things have changed a little. The car used to be "slow" all of the time. Now it's a little bit faster the first 15 minutes I drive the car. The difference is very noticeable, and is also reflected in my 0-60 times. The first 15 minutes or so, my 0-60 times are 8.8 seconds average. After about 15 minutes, the car suddenly looses HP at higher RPMs. My 0-60 times drop to 9.4s-9.6s again.

On the highway, I also get a code 32 around the same time it looses power. But I don't think the two are related. This is why I think so: I got a tip from Oreif to test the EGR valve when the car is idling; push the valve up (using gloves of course ) and see what happens. Well, nothing happened actually. It should have stalled but it didn't. So I removed the intake hose, pushed the EGR valve all the way up, and had someone else plug the intake with his hand. The engine died immediately. This told me two things: 1) the EGR was probably "plugged" at its outtake, and 2) there's no vacuum leak (are my conclusions correct?). So I really have no idea what is causing the code 32, I had a couple of spare EGR solenoids, and replaced that and I still got the code 32. Anyway, since I don't think this is related to the power loss, I'll do the code 32 troubleshooting later.

Back to the performance issue. There's one more very strange thing. Very, very, very seldomly (this happened only three times so far), the car seems way faster than normal. The times it happened I didn't have my digital dash in the car so I'm unsure about 0-60 times, but if I had to guess, it was under 8s. It would stay this fast until I turned the ignition off (well duh) but immediately starting the car again wouldn't make it fast again.

Anyway, I have no idea why I don't notice any difference with the new camshaft, or why there's absolutely zero difference between setting the timing to 4° or 11°. I've had several tips so far: changing the timing curve (would need custom ECM PROM for that), making adjustments to the distributor, getting better injectors, change the intake, etc.

First thing I'm going to do is change the exhaust manifolds. I've ported a stock set and will install those next week. Just to make sure the engine is at least breathing a bit better. I don't think that will be the answer to my problems, but it will at least eliminate one known problem.

But then what? What could be causing my problems? Why am I currently not experiencing any performance gain? What could be the source of the after-15-minutes-power-loss?

I have build myself an ALDL cable and have downloaded WinALDL (and everything works), and will do a test-trip with WinALDL next week so what should I look for?

PS, Edelbrock 3790 Specs:

Duration @ .006 lift: Intake 253°, Exhaust 263°
Duration @ .050 lift: Intake 204°, Exhaust 214°
Lift at cam: Intake 0.280", Exhaust 0.295"
Lift at valve: Intake 0.420", Exhaust 0.442"

[This message has been edited by Cliff Pennock (edited 08-09-2003).]

Hey all you cam guys. Doesn't that cam look like a nitrous grind cam. With all my research over the past few months regarding nitrous and stuff. That is very similar to a nitrous grind. A shorter intake duration and lift. But a larger exhaust lift and duration. It is only slightly larger intake than a stock cam but the exhaust is pretty big. Cliff, I dont know a whole lot about your car. But that cam needs a high flowing exhaust system to run right.

Hell looking at the intake duration, it might even be very close to stock. I would be interested in knowing what the overlap is. Probably idles like a kitten with cat-nip. The only thing short of swapping cams that I can think of is installing larger rockers. IE 1.6 roller tip. Or just toss a 50hp nitrous shot at it. HeHe. I dont have any 0-60 specs for my car with the nitrous but 0-75mph is about 9 seconds.

------------------
85GT 2.9 4spd MSD Everything, Big Cam, No Cat and Nitrous. Nawzz babeee!!!!
http://www.captfiero.com

Cliff,
Did you get a chance to verify the vacuum hose routing that was mentioned in Phil's thread? The performance and EGR could be related if the EGR solenoid is hooked to full vacuum and the MAP sensor is hooked to ported vacuum.

Thread: https://www.fiero.nl/forum/Forum2/HTML/039173.html

Here is what I posted in the other thread:
Judging by your post, It sounds like the "20 minute" thing is a constant. (You can get the car to do it often).
Just out of curiosity, try this. Drive the car for the 20 minutes, When it acts up, Turn the car off and short out the ALDL connector, start the car and drive it. (The check engine light will flash) If the car seems slow right away, Then the problem could be something is telling the computer to go into "limp mode" (Open loop) This means a sensor is bad and the computer is not going off the feedback from the engine.
It could be the ECM temp sensor is bad. With the 20-minute thing, it sounds like something is expecting the car to "warm-up" but it never gets the signal so it "times out". The Temp sensor sticks out horizontally from the intake manifold just below the thermostat housing.

As for your EGR code 32, When they had the engine apart, did they hook the vacuum lines up to the right locations? If they hooked the vacuum from the solenoid to full vacuum instead of ported vacuum, that could cause your solenoid to always be "on". HMM, I wonder if they swapped the ported vacuum and the full vacuum lines from the MAP sensor and the EGR solenoid??? The MAP sensor should have full vacuum attached to it to read the pressure, If ported vacuum was attached it could cause some strange fuel management readings. You could check them with a vacuum gauge. I'll see if I can locate a vacuum routing drawing. There is usually on the decklid but I know mine was removed when I had my decklid redone.
If they were "reversed" this could be the answer to both problems.

Here is a vacuum routing drawing from the other thread:

------------------

Happiness isn't around the corner...
Happiness IS the corner.

Cliff, the change in behaviour after warmup is probably due to the car going into closed loop status where it may be reacting to a problem that it ignores in open loop.

In any case, the test you did with the EGR tells me that something is wrong with the EGR system. The engine at idle should run roughly when you manually open the EGR valve, and return to normal once the valve is allowed to close again. Since your engine didn't react that way, the EGR passage is probably blocked as you surmise.

What does this mean? Well, the ECM most likely subtracts fuel when EGR is being commanded, this is because the O2 in the recycyled gasses has already been depleted. So, if you have a blocked EGR system, the net result will be some leaning out of the mixture. The O2 sensor will pick this up and the ECM will attempt to compensate, but the net result is a loss of efficiency.

I think that you need to correct the problem with the EGR system before proceeding. It obviously has a problem, and that just throws more variables into the mix. How is a code 32 set? Well, it's easier than you might think. On the diagram posted in this thread you will see an item labeled "Vacuum switch" that is bolted to the EGR solenoid assembly. Whenever the ECM tells the solenoid to provide vacuum to the EGR valve, this vacuum is detected by the switch which is plumbed into the vacuum line that goes to the valve proper. If this switch isn't tripped, either because there's no vacuum or because the switch or it's wiring is bad, the ECM sees this and sets a code 32. That's it. So to properly diagnose this, all you have to do is this:

1. Make sure that the vacuum lines are intact and routed correctly, see diagram. Plug the ends and use a hand-held vacuum pump to check the lines for leaks.

2. Check that the switch operates in the presence of vacuum, you'll need a meter and a hand-held vacuum pump for this.

3. Check that vacuum actually opens the valve. Use the pump on the EGR valve nipple and see that it a)operates and b) holds vacuum.

Back to the lack of EGR flow, it's likely that the valve is carboned up. You can remove it and clean the carbon out with a screwdriver and some pieces of wire. Also clean the carbon out of the crossover pipe if any, and clean the carbon out of the EGR tube. This would be a good time to check for cracks in the tube, BTW.

On the back of the solenoid, the end that faces toward the front of the car, there is a hose that runs to a metal tube on the firewall. Make sure that hose isn't blocked. That hose provides the air that the solenoid uses to bleed off the vacuum to the EGR valve. If it's blocked, the EGR valve won't close and if the ECM monitors the vacuum switch for lack of vacuum when there should be no vacuum, that too will set a code.

Whew! I didn't think this would be this long, but to sum it up, you need to deal with known malfunctions first, otherwise you're just chasing too many variables.

JazzMan

I like the specs of that cam. It's a little less duration than the Crane with a little more lift. That should give you a nice torque boost in your mid range. Definitely better specs than stock. Sounds like you have some other issues to work out first.

the fact it seems thermal is very interesting.... It's not just a closed loop problem. Most engines go into closed loop within a few minutes. The probem would manifest much sooner than 15-20 minutes.

Many engines go closed as soon as the O2 sensor is running regardless of engine temp. I think the V6 Fiero programming might also want coolant temp above about 160-170F. An ECM scanner will tell you exactly when it goes into closed mode.

It usually doesn't take 15-20 minutes for any engine that is running right and has a good thermostat to reach full coolant temp. My bloody little L4 is making full heat inside 3 miles even in very cold weather.

There are allot of other things that can take that long to heat up... ignition parts, fuel system parts, wiring, etc. You might want to check fuel pressure after it has run awhile.

Build yourself a cable and run WinALDL on a PC or laptop. This can be helpful, especially when used with other tools...

Just as example: find a pyrometer, preferably a thermo couple type, and check the temperature of the intake air and the coolant. If the ECM numbers aren't close to what the thermo couple reads then you've got a bad sensor or batty wire. Either temp sensor messed up will screw up performance. (Note, If you have a Fluke or other standardized multimeter you can get temperature module and at least one K type thermo couple fairly cheap. www.fluke.com to see an example.)

Check all the grounds and read Wire Service in my cave. Just one bad ground or other wire can reek havock on ECM signals.

------------------
No good ever came from dark and spooky. Norville "Shaggy" Rogers

The Ogre's Fiero Cave (It's also at the top of every forum page...)

Ok, great advice everyone. I'm going to install the ported exhaust manifolds on monday. The trip to the garage is about 30 minutes, so that would be a good trip to log WinALDL data. I'll need to set the ECM in "ALDL mode" (by jumpering terminal A&B with a 10K resistor), right?

I'll check to see if the vacuum lines are connected properly. I won't do any repairs to the EGR just yet, because I want to know what the performance difference is after installing the ported exhaust.

I'll post the WinALDL data here.

quote Originally posted by Cliff Pennock: Oh, one more question: it isn't a problem driving around with the 10K ALDL jumper, right?

Nope, not at all.

JazzMan

Cliff,
I've had this problem several times with my 86GT and 86 SE, both stock and it ended up being a bad EGR solenoid, had a hairline crack in it on the GT and a cracked vacuum line on the SE. Both set a code 32 and acted very much like you described

------------------

Well, dress me up and call me Shirley.

I went to the garage today with my ported exhaust manifolds, thinking it would take them 2 hours tops to put them on. Turns out it'll take them 10-11 hours becaue they want to completely remove the engine. Costs will be around $500! Sorry, too rich for my blood.

I wish I had a garage so I can do this stuff myself...

Anyway, so the manifolds have to wait. I don't think I want to spend $500 on something which will give me 8hp tops.

At least I've got 40 minutes of WinALDL data to analyse, maybe that will tell me what the problem is.

I think the sensor/Vacuum problem are good points. One thing you could do is buy a set of Rodneys steel Vacuum lines. You would have to take off the upper plenum, TB, EGR tube, but all in all not a bad job.
Also, I would check cold and hot compression on each cylinder. I have seen "rebuilt" engines that are truly way out of spec.
Cliff, I can't believe you don't have a friend/relative with a garage? How much do you trust the mechanics that did this job?........Paul

Garages aren't all that common here.

Forgot to say that I checked the vacuum lines and they are all hooked up properly.

And then, there was silence...

the MAP sensor is gonna be pretty bouncy - its linked with throttle opening - overlay a TPS with the MAP. You'll see a small TPS change makes a pretty good MAP change. Tho I dont know if the change should be THAT wide. The O2 sensor - thats the way the ECM does it - slams back -n- forth on the mix. thats why most people who have Air-Fuel guages will say that they are generally useless because they just go back-n-forth. if it stays slammed on one side or the other, thats when theres a problem.

edit - oh yeah, can you add a CTS reading with the last table?

[This message has been edited by Pyrthian (edited 08-12-2003).]

Sure.

Blue = MPH (min: 0mph, max: 97mph), Green = RICH FLAG (min: OFF, max: ON), Red = CTS (min: 74.8, max: 200.3)

looks like this may be the openloop/closedloop issue as mentioned above. That motor didnt go into closed loop until you came to a stop after that speed run. Is that also when it started running worse? I dunno what the time frame is there, but it looks like at least 10 mins before it got to temp. so, may also want to change your thermostat, but then again maybe not - when its warm its slow...! or - completely remove the thermostat and keep it from going closed loop - but thats not a fix, thats a cover up....

[This message has been edited by Pyrthian (edited 08-12-2003).]

quote Originally posted by Pyrthian: or - completely remove the thermostat and keep it from going closed loop - but thats not a fix, thats a cover up....

Actually, the thermostat is removed. I always remove the thermostat in the summer and put it back in in the winter.

Also, according to the following graph, it goes into closed loop when it's supposed to:

Blue = MPH (min: 0mph, max: 97mph), Red = Closed Loop (min: OPEN, max: CLOSED)

Edited for spelling

[This message has been edited by Cliff Pennock (edited 08-12-2003).]

well, I think you need to put a thermostat in that thing, so it can get into closed loop mode, stay there, and get the ECM used to your new motor. also, while searching for a good answer for ya, I came across this: https://www.fiero.nl/forum/Forum2/HTML/035410.html
no answer there, but a clue....

quote Originally posted by Pyrthian: well, I think you need to put a thermostat in that thing, so it can get into closed loop mode, stay there, and get the ECM used to your new motor.

By now I guess I have put about 5000 miles on that engine, so it's hardly a new one anymore.

In the mean time, I have reset the ECM several times and after the "learning period" I have never noticed any difference in performance...

Cliff, some thoughts:

Thermostat, leave it in! The thermostat does many things, all of them good. The thermostat sets the floor operating temp, and the fan sets the ceiling operating temp. The wider the range between the two, or one one or the other goes uncontrolled, the greater the thermal stresses on the engine. If you want to run a lower operating temp, I wouldn't run lower than a 180°, and you should get a lower temp fan switch, say 210° or 200°, temps in Farenheit. The thermostat also provides a restriction of flow through the system, as the water pump pushes against this restriction it raises the block pressure. This in turn reduces the possibility of cavitation in the water pump and the formation of steam pockets in the head, the latter of which is Real Bad (tm). Lower temps will reduce fuel efficiency and power somewhat. On a non-racing motor, lower coolant temps don't really accomplish anything.

The engine will go into closed loop as soon as the ECM can get a useable reading from the O2 sensor, pretty much regardless of what the coolant temperature is. The coolant temp is used to control the fuel maps, with lower temperatures getting more fuel. This is why fuel mileage goes down with lower coolant temps.

Regarding O2 sensor voltages: The regular style sensor used on our cars has an extremely non-linear response curve, meaning that once you get ever so slightly away from 14.7:1/500mV the amount of voltage change is not proportional to the mixture change at all. In essense, this makes conventional O2 sensors function as little more than rich/lean switches rather than as linear O2 gas sensors. The ECM sees rich, so drives the injectors lean. Then it sees lean, so it drives the injectors rich. It does this cycle several times a second, seeking an average 50/50 split between rich and lean output of the sensor. What does this mean? Well, what you would be looking for in a O2 sensor reading is a constant swinging back and forth reading, at least 2-3 times a second.

Regarding WinALDL data: The baud rate on the V6 ECM's data pin is slow, agonizingly slow. It is useless for really monitoring the O2 sensor voltage because the sensor is changing faster than the baud rate. To accurately monitor the sensor will require a meter with a fast-acting digital "needle" like the Fluke 87088 series meters. Do not attempt to measure the sensor directly with a meter that doesn't have at least 1 megaOhm input impedence, better yet 10 megaOhms. You can use an oscilloscope best of all, but they're expensive. The most useable info from WinALDL will be the INT and BLM tables, MALFLGS (error codes), MAP and TPS readings, and the rich/lean counter.

In your graphs above, the vehicle speeds info is pretty irrelevant other than to diagnose a VSS problem. Look at MAP vs O2, but with the X axis spread out some more. Our engines are speed-density, which means the ECM takes the air speed (MAP/TPS) and density (CTS) and calculates the absolute mass of air entering the engine, then supplies the correct amount of fuel for that air mass.

If you want to do a really interesting chart, drive around with the system logging data and the INT tab selected. When the problem starts occuring, hit the "Clear Table" button and continue driving at an assortment of speeds and engine loads (hills, coasting, rpms, that sort of thing) so that you will start filling in cells. It may take quite a few minutes. After a bunch of cells have filled in, hit "Save Table", it will save the table to a text file with the name YYYYMMDD_HHMMSS_INT.TXT. The text file will contain several tab-delimited tables that you can play with in your graphing software. It would be interesting to see a 3-D graph with x and y being the horizontal plane and Z zero being 128, with numbers above this being positive Z and less than 128 being negative Z.

INT is short term fuel trim, the number represent fuel added (>128) or subtracted (<128) by the ECM due to excessive rich or lean conditions.

I think you really should go ahead and fix the EGR system before doing anything else, it adds too many variables to this trouble-shooting process.

JazzMan

The fact it is dropping out of closed loop isn't normal. It should go closed and stay there unless either the coolant temp drops or the O2 goes cold.

Even without the thermostat, the CTS shows enough coolant temp that the engine should stay closed loop. (Yes, you should keep the thermostat in but that doesn't seem to be related to the issue at hand.)

The O2 Sensor voltage is normally all over the place. You worry mainly when it's not doing that. Notice that the O2 sensor is running within a couple miles. That is a good sign.

Also keep in mind that the ECM isn't quite real time. There are often missed changes in rapid feeds like the O2. (I think the ECM only updates the data streem a couple times a second.)

On the other hand the thing drops out of closed loop every time the speed drops. Even during a brief speed drop... The O2 can cool fast but not that fast. At the times the thing drops out of loop, the other charts show enough coolant temp and the O2 sensor still running.

I have to wonder about the wiring.... Like a wire is broken inside the insulation or a connector isn't as tight as it seems. Either of those could be affected by changes in engine position as a result of engine load. They could also be affected by heat. such a flakey connection could be almost anywhere and usually won't be fun to find.

A wiring problem can also account for the R/L flag suddenly going nuts. If the wiring is keeping the O2 from being read corectly then the R/L flag will also likely go nuts just as if the O2 sensor is bad or an injector is jamming.

I also noticed a few spots where the O2 fully is pegged low... I'm not sure thats normal. that starts to happen well before the R/L flag goes nuts.

I don't think there is an injector problem mainly because of the fact it won't stay in closed loop. Again... The fact it goes open loop every time the speed drops a minute doesn't seem quite normal. If that graph is 40 minutes driving, some of those dropouts can't be but seconds long.

Here's a posibility many people over look....

Even if the ECM has been swapped....

It could even be the ECM heating up. ECMs with thermal problems aren't uncommon. I would remove the center console so the ECM can stay cooler and see if that affects if or how long it takes for the problems to show up.

I've done some thermal testing on the ECM and I've seen it go easily to 120-150F in there. (Measured on the ECM cover. ECM internal temp is likely even higher.) This can be more than enough to work a marginal solder joint or connector. 20-30 minutes of driving is enough to drive up the ECM temp, even in fairly cold weather.

Ok, I haven't had time to do some more testing with the car, because I've been working on a program to view the ALDL data in graphical format.

I've zipped up two ALDL datafiles. Both are 30 minute trips, the first one with a cold engine (2003-08-11-Heenreis.txt), and the second was actually the trip back (2003-08-11-Terugreis.txt, a few hours later).

You can get these file zipped here.

To view the data, go to this thread and download and install ALDLView.

Maybe you can have a look at the data and tell me if there's something weird going on...