Car still idles rough/runs rich after 2 months work - I'm at a wall


(Philip Lochner) #81

Greg, if the pickup coil wires to the ignition module is reversed the following will happen (that I can think of):

  1. Timing will be way off - but likely can be “corrected” by rotating the distributor body. This is because the ignition module will now trigger spark from the wrong edge of the “sine wave” that comes from the pickup;
  2. the rotor will not be well aligned with the HT terminals in the dissy and could even jump to the wrong spark plug;
  3. spark timing will not be very stable because the 0-crossing point will be less well defined;

So, yes, it is VERY important to have the pickup correctly connected to the ignition module.


(Gary Crosby 75 XJ6L, 85 XJ-S, 09 XF Supercharged.) #82

Greg,

Have you checked that the harmonic balancer front pulley has not deteriorated? If the rubber has degraded and allowed the outer band (with timing marks) to shift …and you use that to set timing…you may be way off. Just a quick thought…I don’t know what year they added the crank position sensor…so maybe this off target.

Cheers

Gary


(Greg) #83

Thanks for that info, I’m hoping that’s part of it. The new pickup coil I bought even has a + on the red wire connector, so you’d think it’s important. Unfortunately, the Lucas ignition amplifier connector has the blue wire connecting to it instead of the red wire. There is only one way to plug it in (see my previous photo of the plugs). I have bought two connectors and will wire it in such a way where if I plug it in between, I can reverse polarity.

When I last had it running with the red wire going to the blue wire, I timed it to 18 BTDC at 3000 w/ no vacuum advance. But I did notice my timing mark ‘hopped’ a bit at idle and 3000 rpm. It coincided with the engine ‘miss’, and with the timing light missing a flash.

I’ll do spark plugs and compression this weekend. Won’t be able to hook up new ign amp and test spark/timing until following weekend.

I will check harmonic balancer this weekend. That’s a good possibility too. Mine is 88, before CPS.


(Kirbert - author of the Book, former owner of an '83 XJ-S H.E.) #84

It’s not a rise and fall, it’s a reversal – and it’s directional, a reversal the opposite way won’t do it. As the point of a star approaches the pickup, the increasing magnetic flux generates a current in the pickup coil. As the point moves past the pickup and starts moving away, the decreasing magnetic flux generates an equal current with the opposite polarity. The reversal triggers the spark.

If wired backwards, that particular reversal won’t trigger the spark. The reversal that will is the one where the current generated by one point moving away from the pickup is superceded by the current generated by the next point getting closer. When it happens, neither point is all that close, so the current generated is very low, probably bordering on undetectable. When the electronics actually decides a reversal has actually happened could be very inconsistent indeed.


(Greg) #85

what I’m really confused about - if you look at the aftermarket coil pickups for sale, and the aftermarket leads from distributor to ign amplifier, they ALL have the red wire going to the male part of the plug, so it’s impossible to connect red to red, and blue/black to blue/black.

My OEM coil pickup had the red going to the female part of the plug.

What’s going on?

Is there an easy way to know which part of the pickup coil is supposed to be +?


(Kirbert - author of the Book, former owner of an '83 XJ-S H.E.) #86

Sure! Just unplug the connector and connect it up with little jumper wires with the leads reversed. If the car suddenly runs a LOT better, you’ve confirmed the leads were swapped.


(Philip Lochner) #87

PHIXED.:grin:

(asdlfkdf;lkadjfsdfahdljh… to make up the 20 char requirement )


(Richard Dowling, 1979 XJ-S HE+5sp coupe, 1989 XJ-SC, 2003 XJ8 3.5L) #88

Well yes, there is conflicting evidence about which planet is the real one.
Today found my old Leyland workshop manual for the coupe - when it had the original pre HE V12.
That gives the CTS resistance for different temps, a spread of 5.9kohms at 0deg C to 0.19kohms at 100deg C.
Assuming the HE V12 has the same thermistor in the CTS it would have the same spec, and as Kirbert says a paperclip substitute should get you home with a warm engine since it is pretty close to 0.19kohms.

This morning started the convertible with cold engine, but it was a warm night and coolant would be around 20deg C. Pulled the CTS connector and no change.
I looked up the CTS table for the Motec EFI system on the coupe. The Motec manual gives a suggested table of +50% extra fuel at 0deg C tapering off to 0% at 70deg C. I just used that on the coupe.
At a guess the Jaguar ECU has a table not far different to the Motec one.
So if I pull the CTS at 20deg C it will give around 30% more fuel. Hard to say what effect it will have on RPM. The RPM depends primarily on how much air is getting into the engine. Maybe the A/F ratio is still within tolerance to ignite all the air in a cylinder so the RPM hardly changes. There will be unburnt fuel in the exhaust. Pulling the CTS at 20deg C may not be a conclusive test of anything.

I bought the 979 coupe in 1984 and not long after the engine just quit one day while driving home. Phoned the local Jag spares shop and they said good chance the CTS failed, and it was a low cost try anyway. That fixed it.
The CTS must have gone open circuit and the ECU deduced polar conditions and poured in fuel to stall the engine.
Why can I pull the CTS on the stock convertible when it is idling with a warm engine and see no change ?
That looks like an open circuit to the ECU, a very high resistance, and should flood it with fuel.
One thing I did not try was to rev the engine with CTS pulled.


(Greg) #90

also wanted to say, I’ve been brushing up on Timing while I compile my checklist, and if the timing is too retarded, with the spark way too late, that could easily be causing the carbon fouling on my plugs, making me ‘think’ it’s running rich.


(Foggyoo) #91

You may have to add the 18* to the static timing.


(Greg) #92

You mean turn it a tooth on the distributor?


#93

Whew. Whirlwind up in here. All too familiar.

So many things play a role in proper operation or misbehavior in this engine. But if you step back, choose a logical starting point, and proceed to deal with one component at a time it’s not very complex and it isn’t even that long of a list.

If you want a useful guaranteed victory to help your mood and boost morale then find your TDC. Pull spark plug A1 and watch it raise a straw when turning the engine by hand. Stop when it peeks. Check the alignment of the marks on your pully. If they’re off have the damper rebuilt. If they’re aligned buy the good stuff with your damper savings!

Now you have one piton in the side of that mountain. You won’t go backwards from here. And you have proper lubrication for the rest of the list!

Pertinent to the discussion, 4 cyl dizzy, but same waveform.


(Greg) #94

Any words of wisdom on how to hand crank? I know taking out spark plugs will make it easier, but I can’t see how I can get a big socket and breaker in there. From above or below?


(Kirbert - author of the Book, former owner of an '83 XJ-S H.E.) #95

I’d like to hear from others here whether the colors might be reversed on the other side of this connection, the short harness to the ign amp. If that is swapped on your car for some reason, that would explain why none of the aftermarket pickups seem to line up right.


(Kirbert - author of the Book, former owner of an '83 XJ-S H.E.) #96

Gotta admit, I didn’t expect the waveform of the pickup to look like that. Still, if you reverse the wires, you basically flip the waveform upside-down but leave those thresholds right where they are. The timing moves over quite a few degrees for sure.

Fact is, a bunch of this doesn’t jibe with what we’ve heard before about the GM HEI amp. There supposedly is no switching threshold for the coil on; rather, it’s controlled by the IC in the amp based on RPM. At low RPM the coil switching on is delayed to prevent overcurrent in the primary windings. And supposedly the coil off threshold, which is when the spark occurs, happens when the voltage changes polarity – from above the zero line to below on this chart. That’s not far from where they show the threshold, but it’s not the same.

So we now have conflicting info on how the CEI ignition system works. Ah, the wonders of Jaguar ownership!


(Richard Dowling, 1979 XJ-S HE+5sp coupe, 1989 XJ-SC, 2003 XJ8 3.5L) #97

I checked a GM amp unit on the bench a few years back and recorded the results.
For my purpose It was triggered with a square wave.
I was simulating the trigger pulse from the Motec EFI in the coupe.
The Motec unit has a setting for dwell time and that appears to override the dwell time in the GM unit, and maybe that is a design feature of the GM unit. I suspect if the trigger positive pulse is very narrow the GM unit will use its internal logic to set the dwell time and avoid overheating the coil.

Many years ago Motorola made an ignition control IC, MC3334, which was probably used by many manufacturers including GM. The IC provided the triggering function and constant energy ( dwell control ) function. It had on/off threshold 1.8V/1.5V but you could modify that if needed by putting resistors at the input of the IC. Other IC manufacturers may have offered similar items.
You would still need to add a chunky FET or power transistor on the IC output to drive the coil.

The application circuit for the MC3334 shows a DC coupled input and the input pulse from a star wheel inductive pickup must be wired correctly to be a positive pulse to work reliably. You could design the input circuit in front of the IC to be AC coupled and then the polarity of the pulse would not matter. I doubt anybody would bother to add the few extra components for that option.

The attached .pdf file shows the switching waveform. Fairly sure this was posted 3 or 4 years ago when there was discussion on the subject.

LUCAS or GM IGNITION.pdf (257.4 KB)


(Kirbert - author of the Book, former owner of an '83 XJ-S H.E.) #98

No probably about it, that’s what’s in a GM HEI ign amp. I dunno what else is in it. I presume there’s a chunky FET, since the HEI amp requires heat transfer paste under it to keep its cool.

Supposedly Motorola quit making the MC3334 years ago but you can still buy GM HEI amps. I dunno what that means. Perhaps someone just stockpiled a few million MC3334’s. More likely, the newer HEI units are made with some other IC, newer and better.


(Kirbert - author of the Book, former owner of an '83 XJ-S H.E.) #99

That chart shows how the HEI amp works when driven by a Motec unit, evidently producing a square wave. Interesting that apparently the signal never goes negative; it just rises above 1.2V and drops below 0.9V, indicating similar trigger thresholds as the chart showing the Lucas waveform above.

I dunno if it’s germaine to this discussion, but I am curious about the coil-on event. So what we’re saying, I guess, is that the coil is energized when the signal from the pickup (or Motec) rises above 1.2V unless the engine is turning slowly, in which case the MC3334 delays the coil-on event to prevent overheating the coil. Right?

The coil-off (spark) event is apparently clearer now, though. The trigger being caused by voltage reversal is evidently hogwash. It’s actually triggered by the voltage dropping below 0.9V. In the Lucas CEI system it falls well below 0.9V because it actually goes negative, but in the Motec setup it merely drops to 0.


(Greg) #100

Most of that went over my head, but how does it pertain to me if I had the wires reversed and timed to that?


(Kirbert - author of the Book, former owner of an '83 XJ-S H.E.) #101

There are two problems with having the leads to the pickup reversed. First, I’m still not convinced the spark timing will be consistent. With it wired correctly the waveform dives steeply through the threshold. With the wires reversed, it’ll sorta meander past the threshold. If everything is perfect, yeah, it should still be fairly consistent, but if for example that star wheel is just a thou or two off-center it could affect the timing on some cylinders vs. others.

Second, if you had to move the distributor 15 degrees or so to time it, the rotor is now mislocated by 15 degrees relative to the cap when the spark occurs. That’s too far, and may result in some sparks jumping to ground or to the wrong plug.