End of the line?

[Desert Patina]

So, I've had it in the shop for the final gasp of help. The mechanic was patient, and rigorous.

The problem of being gutless remains.

Here's his report.

Compression is at 120 (its better than the 105 I was getting before I got the head machined) and pretty steady across all cylinders.
Altering the cam and timing back and forth doesn't find any additional power, he's found about as best timing as it can have.
Injectors, fuel pressor, all check out.
There is a bit of oil on one of the spark plugs.

He suspects that perhaps the cat is clogged up, and I've chased that phantom before on my 2.3N/A. (it turned out to be over-heating and a major vacuum leak in the brake booster and a slipped tooth on the cam gear all at once).

I don't know that I ever really wanted this to be a complete rebuild. I'm hoping I'm not at that point now.

Anyone know a shop in Phoenix who would be the kind of shop that could take this on at this point?

[Desert Patina]

For the record, I'm still very suspicious of the Cam.

I found this cute setup (Roller cam w/ +95 rockers) http://www.route66hotrodhigh.com/2300Cams.html

[Desert Patina]

Well, the exhaust theory might have something to it. The power drops off at about 5k, and the turbo goes to almost no boost.

[MikeFleming]

What year is the car? Single or dual factory exhaust? Install a test pipe and see if it improves.

Take the CAT section out and see how much air flows through it using a shop vac.

[Desert Patina]

I jumped the gun and took it to an exhaust guy. He said it was flowing fine, and the only curiosity is that he detected sucking and blowing.

I've compiled the list of curiosities.

So here's the straight numbers...

* Even ~120 psi compression on all cylinders
* Code 41,65 when the engine is warm and below 3100rpm
* Near zero boost (if falls off slowly at full throttle) over 4700rpm

Here's the anecdotal:

* The exhaust guy (I just got off the phone with him) says the exhaust flow is both sucking and blowing.
* The mechanic found that the gap on the spark plugs went to almost nothing and he had to re-gap them (i wouldn't mention this except I saw this too and actually replaced the plugs before he saw it)
* The head checked out for no cracks, and was milled
* It has no throttle snap, when idling and put your foot down it accelerates but noticeably slower than you think it would.

[Desert Patina]

For the record, I'm still very suspicious of the Cam.

I found this cute setup (Roller cam w/ +95 rockers) http://www.route66hotrodhigh.com/2300Cams.html

Also for the record, it turns out that Boport and Stinger have both refuted his claims at Turboford. Bo used a simple measurement of the valve travel, as well as measuring the lobe, to find out that both 1995+ and pre 1995 rockers are around 1.8 ratio.

[Desert Patina]

So moving the timing belt a few teeth, and the car is more drivable.

Grappling with some of the underlying issues still...

* you know when you have the car in too low a gear, and it kind of shakes or bucks? Its doing that all the way up to 3k rpm
* Simultaneously, if the car is warm, the check engine light comes on, but goes off around 3k rpm.
* Seeing the same error codes, 41-65 (meaning it can't get out of loop, but wants to because it thinks its lean), the manual says check HEGO circuit.

What I did:

- I replaced the 02 Sensor.
- I verified ground on the 02 sensor by checking resistance between the ground terminal and the ground at the battery (no resistance)

I opened up the ECU connector, and the pin might be pushed up, if so I wonder when my new ECU connector is coming or if locktite might make it work long enough for one more connection.

[Desert Patina]

Found / fixed one more vacuum leak (the PCV hose that goes to the turbo inlet developed a tear) and its driving even better.

Now the codes are 41 (lean) and 42 (rich). Which according to this page still implicates the HEGO circuit. http://www.tbirdheritage.com/turbocoupe/articles/pinpointtests.php?actionstepid=H30

However it also implicates the VAM.

The HEGO and HEGO GND circuits check out to the best of my ability to check.

The driving behavior is still very predictable. The engine light comes on when the car is warm, and in low RPM. I don't know of a circuit failure that is easily associated with that. But that could be a sticking VAM.

The VAM's are plentiful (being in common with a number of BMW models and years). I think I'll just get a remanufactured one and see if that helps.

[MikeFleming]

The engine light comes on when the car is warm, and in low RPM. What is this "Engine Light" you refer to?

[Desert Patina]

In my ever-ready repinning, I repurposed the boost light as the check engine light output from the LA3. It works well, but I had to disconnect the boost buzzer.

[MikeFleming]

We don't need no Stickin' Boost Buzzers!

[Desert Patina]

little known byproduct of the buzzer when it is on for a long time ... it makes the turbo sound like a police siren.

[MikeFleming]

little known byproduct of the buzzer when it is on for a long time ... it makes the turbo sound like a police siren. Cause/effect is backwards. The buzzer doesn't make the turbo noisy, the overboosting does.

[Desert Patina]

Maybe... It was independent of the boost the way we had it.

What the buzzing did was make it sound further away, and masked all but the highest pitch. Going through the gears made it oscillate like a police siren. It was so weird my son was scared we were being pulled over. I didn't see any cops but it still took a few moments to figure out what was going on.

[MikeFleming]

It was so weird my son was scared we were being pulled over. I didn't see any cops but it still took a few moments to figure out what was going on. There was a time on SVO History Past, when folks would pay top dollar to have that noise installed into their cars.

[Desert Patina]

Not willing to let this go, when I'm so close, I've tried a few more things but I'm still at a loss. While I'm away from my car, I'm thinking of what it could be.

1) The driving behavior is exactly the same whether the H02S is plugged in or not. The light will go off above about 3k RPM, even though it is not connected. The codes change from 41,42 to 41, 65 and fault in the H20S circuit. I figure when it goes on it is only when it wants to go into closed loop and can't. When it doesn't expect to be in closed loop, it doesn't give a signal. Does that sound right?

2) The EGR control valve is working, it checks (can't blow through it when off, can when its on). I also learned codes 82,83 are really about the radiator fan normal and high mode respectively. I don't have a fan connected, but I will after the weekend.

3) I also validated the H20S, and I think the only problem is that the heater wire doesn't come on. But I don't think that is required for it to work, it should get the voltage directly from the sensor itself, is this correct?

Chasing down forum posts on many different forums about where a engine has codes for both rich / lean conditions, I've read that the injectors, fuel pressure regulator, and the engine temperature sensor (ECT?) causing the engine to jump in and out of closed loop at the wrong time, are all suspect in these cases. So would a very large cam causing a false lean on an otherwise rich running car. On that note, I happen to be running as best I can tell a smaller early SVO cam on rollers with dual springs.

Any help verifying any of these assumptions will be greatly appreciated. Many thanks.

[Desert Patina]

So how hot should the LA3 let the engine get before kicking off the fan?

I've installed the Volvo fan relay, it works (the fan spins during KOEO), and took it for a drive to warm it up. When I got home, I let it idle for quite a while. I let the temperature get to "L" in "NORMAL" on the dash gauge before turning off the engine.

I took the KOEO codes again and it said "27", meaning it got too hot. But why didn't the LA3 start the fan?

I'm still suspicious of the ECT, now more than the post right before. But still...

[Desert Patina]

I changed the ECT, and quite a bit changed. The car drives much better, and I have the fan working.

Still getting 41, 42, and now 65 too (which is a relief, honestly because it explains the others nicely).

[Desert Patina]

I finally changed the ECU connector, there were two loose pins, and a few (including the H02S) that were over-round.

After the change, only code 41 (engine lean) remains. And while it can be blown out by some WOT, it stays on pretty constantly when the engine is warm.

[update, a second drive pulled up both 41, 42. A third drive only pulled up 42]

[Desert Patina]

We did the paper test on the exhaust, and it was going "thump" kind of like when paper hits the end of a vacuum cleaner hose. It was pretty regular.

So, all this work and changing the cam timing didn't change the blow back. And it is intermittent.

Remember it did this before, while it was getting 120 pounds across all of the cylinders.

I also read on another forum (Facebook?) where the dual springs (which my car happens to have) are problematic.

[Desert Patina]

Actually a few more things that are funny. The Mechanics we called thought the paper test was nonsensical, it simply didn't make sense to them.

Also I had the upper/lower separated and for some reason I seem to be getting a lot of oil through the lower PCV system. It was wet with oil in the #2 and #3 ports (the ones closest to the intake).

[Desert Patina]

Well the last mystery was solved. Unfortunately the 120 compression test was mistaken. The compression is more like 95-100, and the oil test brings it back up to 145.

Oddly enough that is what I got when I first did the compression test, but I thought I was doing it wrong. I only did the compression test, not the oil test.

So maybe it is the end of the line.

I do have, perhaps, a 36k mile motor for probably 1000 shipped. Its tempting, but I've been averse to putting nice stock parts in this car. I feel that a nicer car deserves the nice stock parts.

The other idea is to just get one of the plentiful late Ranger 2.5's for about $300, and swap in the pistons and crank. From what I understand, the only trick is porting over an EDIS from a 90's mustang to be compatible with the LA3.

[MikeFleming]

Not saying you don't have a problem in that cylinder, but don't put much (read: ANY) faith in the "oil test".

Many, many, many decades back the idea was if rings (meaning top ring) were leaking serious compression, adding some oil onto the flat-top or domed-top piston would push the oil to the cylinder walls (at cranking speed) and help provide some thick liquid to help seal the top ring temporarily (we didn't care about those oil smoke clouds as all cars did that some anyway). It would all get burned and washed out the exhaust after a few minutes of running. And least for a couple minutes you'd kill all the mosquitoes in the shop and neighborhood. That was a good thing back in the day. No expensive catalytic thingies to get clogged or overheated.

On the SVO engines (and other like-minded turbo engines of the era) - with their dished pistons, most (>90%) of the injected oil will merely fill the ~12cc depression in the piston and stay there [at cranking speeds]. All this does is reduce the clearance volume in the cylinder so the effective CR goes up proportionally (like from a dished to a flat-top piston). Until the oil gets burned off (which will never happen if it's not started and run for some time).

It's like you added more to the top of the piston - so unless there's a hole in the piston top, the pressure will *always* go up.

[Desert Patina]

So, I've gone from ambiguously floating valves to definitely sticking valves.

As another Bahala-na experiment, I decided to add Neutra #131 to the oil and gas. I bought a gallon of it, might as well use it.

After about 100 miles, it's driving much differently. Before it was sluggish but stable, and the check engine light came on (both lean and rich).

Now, its quite a bit more snappy, but at full throttle I can hear the ignition cycle leaking into the exhaust like a small constant backfire. I can hear the valve clanking as it stutters at a very low rpm launch.

However, the check engine light never comes on anymore.

So I wonder if I over-advanced the cam to mask the floating valves, and now with a cleanup the injectors are firing right and the lifters are all cleaned out and working well (except one) and now it is time to pull the advance back again.

[Raven855]

the manual says check HEGO circuit.

What I did:

- I replaced the 02 Sensor.
- I verified ground on the 02 sensor by checking resistance between the ground terminal and the ground at the battery (no resistance)


Since you are still having rich and lean issues, you may want to check this circuit again. The O2 circuit goes from the computer to the sensor. The sensor ground is actually the ground wire on the turbo and runs to pin 49 at the computer. The signal wire is pin 29. The computer is grounded to the battery through pins 40 and 60. They actually run to the ground lug that is under the battery and not directly to the battery.

[Desert Patina]

Right.

I checked the ground ciurcuit by putting the positive lead into the engine harness connector at 49, and then touched the other to the ground bolt just beneath the ECO. Far less than 5 ohms. I also had someone wiggle the injector harness (for this engine harness, the HEGO Ground goes to a bolt on the intake manifold) to see if that was loose. It wasn't. Everything checked out okay.

Then I checked both ends of pin 29 at the ECU connector and the HEGO connector. Again, no issues. I also replaced the ECU Connector just to make sure the pin wasn't pushing out when I connected it. It wasn't but the VAF pin was, but it is good to make sure.

I also soaked the HEGO connector in citric acid to clean up its connections.

I also put my spare LA3 in, and verified it also saw the same issues.

Today I'm sure I had a sticky injector which was causing it to run rich, and floating valves which caused it to run lean. It switched between both conditions as both were intermittent.

the Neutra has fixed both conditions to where the it runs the same way all the time on all cylinders.

The problem is I don't understand the mechanics of our OHV hydraulics to explain exactly what is happening.

Its like the intake sticks shut occasionally below 2500RPM, and the exhaust sticks open constantly above 3300RPM.

Below 2500, occasionally its more than just a stall, its like the engine revolution runs into something, there is an audible clank, and recovers. To me, I feel like the intake is closed while the piston is moving away from the head, creating a vacuum and slowing the engine down. Then the clank is it finally opening and the car runs as normal (though to what I know about valve train mechanics, that isn't possible but what do I know?)

Above 3500, its like a string of firecrackers were set off in the muffler, only I know it is before the muffler because the turbo spikes really high really quickly.

The Cam timing checks out, its dead on. The head itself was warped (bowed) when I had it checked. So I think probably all the valves are bent also, and the cleaning I just did is causing the bent valves to catch more often (and get more bent in the process).

The last mechanic said its time for an engine rebuild, and I just have to believe him. I'm saving a few thousand to get it rebuilt, or buy someone's old racing motor if it is beyond hope. Probably in about two months.

[MikeFleming]

JUst t ocelar up a few misconceptions.

... the HEGO Ground goes to a bolt on the intake manifold) to see if that was loose. It wasn't. Everything checked out okay. BAD idea. O2 sensors work in millivolts so ANY resistance is really Bad, with a large "B".

There are lots of gaskets and unreliable electrical thingies in between the intake manifold and exhaust manifold/turbo housing. Move that HEGO ground wire to the turbo intake (cold) side somewhere. Much more better.


... floating valves which caused it to run lean. Floating valves won't cause "lean". Lean means less fuel for the same air. "Floating valves" does nothing to the fuel/air ratio; while it does have an effect on cylinder change level.


Its like the intake sticks shut occasionally below 2500RPM, and the exhaust sticks open constantly above 3300RPM. Totally impossible for a valve to stick closed (unless the rocker arm/cam lobe are completely gone and that's a different issue). If any are sticking open, then the valve springs are gone.


Below 2500, occasionally its more than just a stall, its like the engine revolution runs into something, there is an audible clank, and recovers. To me, I feel like the intake is closed while the piston is moving away from the head, creating a vacuum and slowing the engine down. Then the clank is it finally opening and the car runs as normal (though to what I know about valve train mechanics, that isn't possible but what do I know?) This makes absolutely no sense to me.

Clank like pieces of piston floating around?


... because the turbo spikes really high really quickly. Elaborate what "turbo spikes" means.

What is the history on the valve springs?

[Desert Patina]

Thanks for clearing that up.

FWIW, the ground on the intake manifold is Ford's idea. It came with the 89 harness setup.

Just to be clear, because there are two grounds at work. One ground is from the ECU, and one ground from the HEGO. The HEGO itself is grounded directly to the negative battery terminal, and through a connector (Again this is the '89 harness). I believe that simply grounds the heater.

The HEGO GROUND circuit (49 on ECU) that is very sensitive, which usually is the orange wire on the Turbo, on the 89 harness goes directly to that ground strap on the intake manifold to ground directly to the block.

Again, not my idea. I figured Ford thought it was an improvement. My actual 89 doesn't seem to have any issues with that configuration, though I have no doubt what you describe would be better.

As usual, I take your word for all the rest also. But to describe my setup better...

The low RPM clank happens after the stall, as if it is more freeing something up than hitting it to begin with. Like a running into a wall with a balloon, first you slow down then there is a 'pop' and you go the rest of the way to the wall pretty easily.

The pistons have been inspected quite a few times as best they can be through the spark plug hole, and they are whole. Besides, it isn't a clanking, it is a one time clank only after the engine seems to stop.

The turbo spike happens when the quick succession of backfires happen. The boost gauge very quickly goes up to 22+ where before the fire-cracker sound it hangs steady at 8-10psi.

The "lean" I worry is like a false lean that some cars get from a big cam. Floating valves, as far as I can tell, act like a high duration cam. Besides, like I said before, the light doesn't even come on anymore.

And again, I have no idea. All I can figure at this stage is that the engine and head need a rebuild.

[Desert Patina]

Oh, and the history of the valve springs is a mystery to me also. I bought the car this way, and its a bizarre setup.

Much of what the head is, is a mystery to me.

I did identify the cam as likely a smaller early SVO cam.

Here's the measurements I did with a caliper on the intake and exhaust lobes...

High - Low
I: 1.40 - 1.18
E: 1.41 - 1.18

The stamp on the cam was in the back, around the circumference of the shaft itself, and said, 19950696b1a

It has double springs and a roller.

Which means, it is a slider cam with a roller. And the TurboFord folks seem to take a dim view to double springs.

Thanks again for your help.

[MikeFleming]

It has double springs and a roller.

Which means, it is a slider cam with a roller. NOTHING good will ever come from using roller followers on a slider cam.

REALLY BAD COMBINATION.

[Desert Patina]

Agreed. I called Jerry at Shneider Cams a few weeks ago when I found out, we'll work something out when it gets rebuilt.

[Raven855]

The O2 sensor started as a single wire sensor with the ground or return supplied throu the sensor casing. It was not Fords idea but the way Bosch implemented it. The first cars to get it were Chryslers in the mid 70s. There were 2 problems with this design. The first being the potential for fluctuating resistance affecting readings. The second was the time it took for the sensor to get warm enough to work. They solved the second problem by adding a heater to the sensor. This is the 3 wire sensor we have on the 84 and early 85 cars. In late 85 and 86 they reverted back to a single wire because of the lack of wiring space to the computer. Eventually a forth wire was added to the sensor on later cars. This is a dedicated ground for the sensor return. The 2 wires for the sensor(not the power circuit)in many cases is shielded wire. Mike pointed out that the sensor reads in millivolts. It is susceptible to electrical noise, especially from the heater circuit of the sensor. The egr circuit, injector drivers, bcs, grounds and alternator can wreck havoc with readings just so you know.

From Bosch, oxygen sensors can fail when the sensor’s ceramic element is exposed to certain types of silicone compounds or when an oil-burning engine leads to the sensor becoming oil-fouled. Also, a small amount of poorly refined gasoline can kill an oxygen sensor, as can the use of some over-the-counter fuel additives which are not “oxygen sensor safe.”
These are the primary reasons for early sensor failure. Also don't use an analog meter to check the sensor, you could ruin it. Use a digital meter or scope.