Installing New Injectors

[RRoller123]

Hopefully this will finally solve the rough idle problem. New injectors are going in. I had the original ones cleaned about 8 years ago, and I know for a fact that they are at least 21 years old, so it seems like it is time. Using an American made set from Allison's that I have had sitting around in sealed bags for a long time.

The engine has classic injector leakage symptoms, like engine shake, rough idle, won't restart (flooded), but runs very well at open throttle and after warmed up. Just won't idle well below about 1300 RPMs. All sensors, etc checked, new short block, head leak down test, compression good, all that stuff is ok. Fuel pressure checked, fine at about 36 psig. Not a sensor problem, most are either new or checked and tuned. Starts fine when cold, and runs great for about 7 seconds, so that tells me a lot of things are working right (TTS, CSI). But then just runs like gumpcukruss until it is fully warmed up. After that, idles ok, but not below about 1200-1300 RPMs. And still has some vibration, even after the new short block was balanced with the clutch assy.

Engine also has a tendency to flood when allowed to sit for more than, say 15 minutes. When that happens, I can solve it by pulling the air tube and using a vacuum cleaner to exhaust the gas fumes. Then it starts right up.

When I pulled the old injectors out, all of them had slightly wet tips. And the car hasn't been started or fuel pressure powered up in days. Something tells me that is not right...

[18Fiatsandcounting]

The engine has classic injector leakage symptoms, like engine shake, rough idle, won't restart (flooded), but runs very well at open throttle and after warmed up. Just won't idle well below about 1300 RPMs.

My memory is fuzzy, but I thought you put performance cams in this engine? Don't these cams cause some of the symptoms that you're describing?

-Bryan

[RRoller123]

Yes, but not anywhere near as bad as these symptoms, and the cams are 274FI, relatively mild. The symptoms were there with the stock cams too. The radically lightened flywheel seemed to have more of an effect on the idle quality.

Plus the flooding has no other explanation, nor does the wet nozzles on the injectors. It is interesting to note that one would expect the injectors to leak at the tip where the pin seals the orifice, but the wetness is around where the nozzle crimps to the injector body, not at the nozzle pin orifice. Knowing the POs, I bet these are original, 41 year old injectors. Time for a change.

[18Fiatsandcounting]

Yes, but not anywhere near as bad as these symptoms, and the cams are 274FI, relatively mild. The symptoms were there with the stock cams too. The radically lightened flywheel seemed to have more of an effect on the idle quality.

Thanks for the clarification. I have only one experience with a "lightened" flywheel, and that was decades ago when I forgot to tighten the 6 bolts that hold the flywheel to the end of the crankshaft. Eventually, the bolts worked loose, and one morning the engine started but then the bolts were totally loose, so the engine was running but with no flywheel attached. I do remember how impressive the engine revving was with no flywheel weight! I've never seen a tach needle move that fast...

Unfortunately, my knowledge of fuel injectors is extremely limited so I can't offer much of value.

-Bryan

[RRoller123]

I am no injector expert either, but am gaining knowledge quickly!

[micbrody]

I’m sure you have gone through all the stuff- checking for air leaks; measuring resistance on temp sensors; AFM measurements; etc.

I had a problem where car always starts when cold. But after stopping car to go pick up something from Target, the car had a huge difficulty starting back up.

I also had checked everything I could think of. And I also visually saw csi squirt at start up when cold.
However: I never checked to see what csi did when engine hot. Turned out the TTS was always stuck to “continuity”; so CSI always squirted during start up- regardless of engine temp. So it flooded engine when hot.
When I switched TTS, it has started beautifully EVERY single time.

I would re-investigate CSI/TTS. From your post:

“Starts fine when cold, and runs great for about 7 seconds, so that tells me a lot of things are working right (TTS, CSI)”

“Engine also has a tendency to flood when allowed to sit for more than, say 15 minutes. When that happens, I can solve it by pulling the air tube and using a vacuum cleaner to exhaust the gas fumes. Then it starts right up. ““

Maybe CSI is stuck open- always squirting; even during normal running of engine. (Mine was not stuck- but TTS was stuck to energize- so during starter sequence, it was always squirting, but would stop after starter energy stopped(release of key from “start” position).
Since you seem to have problems after car is running, maybe CSI mechanically stuck open.

Try this: after car is warm. Shut off car. Clamp off fuel line to CSI (or disconnect and plug up both fuel line and hole for CSI. Then restart car and see if things run better

[RRoller123]

Good experience, Thanks. I will try that! But I do have a brand new Bosch CSI in there that has maybe 1500 miles at most on it. Yet that is a good idea, as we tend to forget that there are FIVE injectors that can be leaking, not just 4...... So if the change of injectors doesn't completely solve the problem, I will do that test.

Other stuff to do: Also, I have a new AAV (but a few years old) and will clamp off the feed line from that too. Although I have visually checked it and it opens and closes properly, it has an odd overlapping crescent moon baffle internal design, that we can't be sure is actually closing. I have a couple of old ones that I could easily swap in and see if it changes anything. The AAV is adjustable, by the way.

Another possible issue is that the charcoal canister may be a problem. I noticed some gas smell in the trunk area the other day, and can only surmise that fume pressure was not relieving through the canister to atmosphere as it is designed to, since the gas cap on FI systems is SEALED and pressure from solar heating, filling the tank, etc vents back through the charcoal canister. (readers: We should all pay attention to this, there are 2 different gas caps: Sealed for FI, a pressurized system, and a Vented cap for carb systems. It is really easy to get the wrong cap on there, they both will fit either system), The pressure was somehow forcing fumes back through the fill cap or tank seal, etc. (The "2-way valve" that is so misunderstood, is only in there to maintain a slightly above atmospheric pressure in the system, on both sides, hence 2-way.)

The canister also has a continuous connection by intake vacuum to the intake manifold to burn off collected hydrocarbons, so a screw up there could cause excess un-metered air. I suspect that this is an original charcoal canister, and although I HAVE the notorious missing metering cap, and it shows a good vacuum pull when running, and I cleaned out the gumpcukruss up top under the cap, the charcoal itself may be pooched and clogged, causing flow problems. I really like this particular pollution control system, it works really well and gives excellent environmental benefit for a very minimal if any performance cost, and also involves very little mechanical equipment to work properly. So I will keep it working properly.

Same with brake booster hose, possible un-metered air source. So best thing is to clamp off all of these potential un-metered air sources, one at a time, and see what the effect is.

Lots of stuff to keep one busy......

[maxm50]

Great to see your progress, Pete! Sounds like the new injectors should make a big difference, if the old ones have wet tips... Sounds like they are slowly leaking due to the residual pressure in the feed lines.

Regarding the charcoal canister cap - I just bought one from Autoricambi that they had laying around (because I had read that it's a metered orifice), and I can confidently say that it is NOT a "metered orifice" as has been claimed on the forum. It is strictly to prevent water from entering the canister. It's just a little plastic umbrella cap with PLENTY of open airflow around and underneath it. I'm a mechanical engineer and Fluids class isn't TOO far back in my memory, and this cap certainly wasn't designed by anyone intending it to meter airflow. So, don't worry about it. When I get some free time over here I'll 3D print a bunch of copies of it and make those available to the forum.

I am curious about refreshing an old charcoal canister, though. How do you plan to do that?

You can test to see if your AAV is closing fully by leaving it hooked up to a 12V battery for a few minutes. You probably knew that. It'll close slower than it does on a running engine, but at least mine closed fully with only 12V power. Not sure if that's normal or not.

The CSI malfunction that micbrody mentions is a great idea. But I wouldn't look at the CSI - I'd look at the thermo-time sensor temperature sensor that signals it to open. If the injector were stuck open, you'd be getting fuel from it all the time, because it always has fuel pressure - not good, and doesn't sound like that's happening to you. If the thermo-time sensor is faulty, I think you'd get symptoms more like you're describing. But I'm still learning this stuff. Should be very easy to test: just unplug the electrical connector from the CSI after you've warmed the engine, then stop it and try to restart, see if it starts up better. If it does, then I'd say the thermo-time sensor is bad.

Lastly, as you know I've been fighting similar idle issues on an FI motor. My AFM had definitely been monkeyed with by a PO (plug on emissions screw was missing), and my resistances were a bit wonky. I had read many things about bending arm contacts, adjusting spring tension, setting screw via O2 sensor voltage, etc etc. I ended up deciding that for $250 + shipping, I'd rather have a professional rebuild it with new parts, not just bent springs. So I sent it to the folks at www.fuelinjectioncorp.com, and it returned yesterday looking like a brand new part. I'm happy to say that it solved the remainder of my idle issues, and it feels like a new car now. So, something to think about if you've exhausted other options! In the grand scheme of things, it was money VERY well spent...

Actually, one more thing - did you install those throttle shaft seals I sent? Curious to hear if they made any improvement!

[RRoller123]

I did install the throttle shaft seals. It didn't have an effect on the idle issue, but I can say that the throttle shaft definitely fits more securely. It is clear that the old seals were sloppy in there. The shaft has no play at all now and rotates with just the slightest bit more resistance, which is what one would expect.

The charcoal cap, I agree doesn't do much metering, the opening is just too large, There is a long series of posts somewhere here on the Forum about opening and refreshing the canister with aquarium charcoal, but I think I would just buy a new one if I find that it is clogged. That is also coming up shortly on the diagnostic test list.

I am glad to hear that the AFM rebuild worked!! Fantastic!. Mine was dragging and I did the flap sanding procedure, which helped. But the bypass has obviously been screwed with in the past, and the thick film resistive material has worn tracks. I did the arm bending process and that helped too, but I may have reached the limit of what can be done at the home workbench.
That would be my next step if the new injectors don't resolve the issue.

I think the bottom line is that nothing in the FI system is intended to last over 40 years, and the FI cars will need to have literally everything either replaced or rebuilt at some time. The carbs are certainly a lot simpler!

[RRoller123]

Well it definitely solved the restart problem, so that is a big success! I am certain that the engine was flooding from leaky injectors. Car restarts instantly after a 1/2 hour set from full warm.

When the fuel pump comes on (I have it wired to the ignition "ON" position), there is no slight gurgling sound, as there was when it had to purge the small amount of air bubbles in the old injectors that had come in from their slight leakage, and was clearly flooding the engine. So now the car will reliably restart (I hope, could always be something else that comes up) and I will be more confident taking it out to cruises, etc where the engine gets shut off for awhile.

But the rough idle issue is still there. Going after the AFM next. But the car runs well, I found that the ignition timing was off a bit from the necessary high idle (was somewhat retarded), so played around with that. I can get the car to idle a little lower now, maybe 1000 on the tach, but who knows whether that is a real indicator? I have a tach gauge that fires off of a spark plug, installs like a timing light pickup, and then indicates "true" RPM based upon the number of cylinders in the engine. Will try that out and report. Curious to see how accurate the tachometer may be.

Always something to play with and explore with these little cars.

[RRoller123]

I am curious about the proper position of the spring stop, in the AFM, for the sweep armature.

In my AFM, which has been messed with a lot by both PO's and myself, the spring that stops the armature is SET BEHIND the stop bar, like this:



This lets the armature (i.e. the AFM door flap), slam hard when it closes. It also sets the position of the sweep arm almost completely off the thick film resistive material field? Somehow doesn't seem right.

The spring could also be set IN FRONT of the stop bar, which keeps the sweep arm a little ways out on the thick film resistor material, and keeps it from slamming hard closed when you take your foot off the pedal. Like this:



Anyone know which position is correct? They both would work. THE SPRING SET BEHIND THE STOP BAR closes the door fully, but it slams hard. I am wondering if this is why the cars tend to drop to almost a stall condition when you take your foot off the pedal, then recover quickly?

With THE SPRING SET IN FRONT OF THE STOP BAR, the door has a soft close, it doesn't fully shut, and the sweep arm stays out on the resistive field.

Wonder which is correct? Or rather, which is factory?

[tima01864]

intuitively the top photo position is correct, The bottom position photo would just not let the spring do its thing.
Looking at my old AFM the post holding the spring is 90 degrees from the front casting andparrelel with the side. Yours looks bent to the right

[RRoller123]

Very odd, I wish you could figure out how to post pictures. I am not sure what is different. You mean if that small bar at the center of the coil spring was rotated 90 degrees CW it would match yours?

Maybe that is how that small spring tension is adjusted? All this thing does is stop the armature from slamming hard against the stop when I take my foot off the pedal. On mine, it stops pretty far back from that vertical stop post, which puts the swing armature almost off the side of the resistive material. Wondering if this is a source of the idle problems? So I am going to clean everything up with isopropyl today and try it in that other position and see how it runs.

[tima01864]

Yes if the spring post is rotated say 45 gegrees or in a 90 degree position it woul match the one i have. The spring is not strong enough to add resistance.

[RRoller123]

Not sure what you mean that the spring is not strong enough to add resistance? Electrical resistance? It has enough tension to stop the armature from slamming against the hard stop just fine, but it settles in at a point pretty far back from the post and the armature is then almost off the circuit board. I have a feeling that this is not right?

Yours has the spring mount post at 90 degrees to the case, mine is shifted, maybe this is an adjustment of sorts. I will fiddle with that today sometime. Or maybe not.

Because this is a historic day here on the Canal! The Mayflower is coming back to Plymouth, after a 3 year full restoration, and will be coming through the canal in about an hour! Off with the camera we go.