Very perplexing no start/spark issue.

[AustinK]

Two new sensors didn't fix the problem. Still no spark. I tested their vac after install. Here's what one of the test procedures said...

Pull rubber boots back off the two sensor plugs. Pull the main relay & fuel pump relay to disable the FI system.
2. Set your voltmeter to AC volts and back probe the black and yellow wires while cranking.

Speed Sensor (Black Connector) -- 4.0 AC volts max at crank.
Ref. Sensor (Grey Connector) ------ 0.4 AC volts max at crank.


On my speed sensor I got over 4 vac. Almost 5. Could this mean something since the procedure says 4vac max? The reference sensor was under the 0.4 vac.

[AustinK]

I've been cross referencing some of the Porsche motronic articles and they talk about tach bounce when starting as in indicator of good speed and reference sensors. Do our tachs bounce the same way?

[slofut]

I feel your pain... Don't give up though, it'll be such a blast to drive!

[Brucey]

I've been cross referencing some of the Porsche motronic articles and they talk about tach bounce when starting as in indicator of good speed and reference sensors. Do our tachs bounce the same way?

even if they do, this falls into the 'necessary but not sufficient' category.

Such tests (that are easy to do but are not unequivocal indications of 'engine health') can indicate a fault condition exists, but if they appear to 'show good' that doesn't mean that the parts in question are definitely working correctly. This is OK for a 'screening test' because if you do ten simple tests of this sort and you find an apparently faulty part then it is odds on that this part is what has just failed.

However such tests are usually of no value whatsoever when it comes to discriminating between parts that are good and ones that are only half-good. You clearly have a fault and the reason you have not seen it yet is almost certainly because you keep doing tests that are not good enough. Doing more tests of this sort will give you clues at best and at worst will see you running round in circles chasing your tail for weeks.

That is why I have suggested that you need to carry out proper tests (eg -as described in the workshop manual) that are capable of identifying parts that seem to be working but in fact are not quite good enough.

cheers

[AustinK]

I've been cross referencing some of the Porsche motronic articles and they talk about tach bounce when starting as in indicator of good speed and reference sensors. Do our tachs bounce the same way?

even if they do, this falls into the 'necessary but not sufficient' category.

Such tests (that are easy to do but are not unequivocal indications of 'engine health') can indicate a fault condition exists, but if they appear to 'show good' that doesn't mean that the parts in question are definitely working correctly. This is OK for a 'screening test' because if you do ten simple tests of this sort and you find an apparently faulty part then it is odds on that this part is what has just failed.

However such tests are usually of no value whatsoever when it comes to discriminating between parts that are good and ones that are only half-good. You clearly have a fault and the reason you have not seen it yet is almost certainly because you keep doing tests that are not good enough. Doing more tests of this sort will give you clues at best and at worst will see you running round in circles chasing your tail for weeks.

That is why I have suggested that you need to carry out proper tests (eg -as described in the workshop manual) that are capable of identifying parts that seem to be working but in fact are not quite good enough.

cheers

Thanks Brucey. I've got the new speed and reference sensors, new coil, a know good DME. So am I into the wiring harness at this point. The reference sensor's wire from connector to pin out at the harness held up under load. The only wire I didn't know how to test under load was the reference sensor ground. Any idea how to do that? I can't put 12 Vdc on it, can I?

And thanks for being patient. I really really appreciate the help.

[Brucey]

the other two wires are the important ones in the speed/ref sensors; the ground is a screen to prevent interference. Unlike the other wires if it fails it probably won't give a no-spark condition. For this one you can test using a decent multi-meter; its 'load' is only ever small.

BTW there are some favourites for causing no-spark, including;

- speed and ref sensors swapped over (where they plug into the harness)
- main and fuel pump relays swapped over
- that wire from the starter to the ECU (that gives a start signal to the ECU) doing the wrong thing. But there are plenty more...

you are the guy that needs the patience BTW; it is easy to get frustrated, or start playing 'parts-change-bingo'. Keep at it, I'm sure you will get there in the end.

cheers

[songzunhuang]

Oh boy, it's my turn now.

After driving along quite well for a week and over 150 miles, this morning my car wouldn't start. It was fine when I drove it home and parked it last night! I guess I should b thankful that it died while parked in my driveway.

The reason I am posting in this thread is that I did a couple of quick checks and I do not seem to be getting spark anymore. I am now looking for clues. Here's what I've done so far.

1. Jumpered the main and fuel pump relays. Doesn't seem to be the issue. I hear the fuel pump kicking in when jumpered.
2. Checked spark at plug #5. Nothing.
3. Check spark coming off of main coil wire. Nothing.

I had to get to work at this point so I just jumped in another car. Now I can't stop thinking about even though I am at work. What shall I check next?

Wire from started to ECU? I'll have to dig into that more. I'm not familiar with this one. Sigh, more things to look over this weekend.

[slofut]

Oh boy, it's my turn now.

After driving along quite well for a week and over 150 miles, this morning my car wouldn't start. It was fine when I drove it home and parked it last night! I guess I should b thankful that it died while parked in my driveway.

The reason I am posting in this thread is that I did a couple of quick checks and I do not seem to be getting spark anymore. I am now looking for clues. Here's what I've done so far.

1. Jumpered the main and fuel pump relays. Doesn't seem to be the issue. I hear the fuel pump kicking in when jumpered.
2. Checked spark at plug #5. Nothing.
3. Check spark coming off of main coil wire. Nothing.

I had to get to work at this point so I just jumped in another car. Now I can't stop thinking about even though I am at work. What shall I check next?

Wire from started to ECU? I'll have to dig into that more. I'm not familiar with this one. Sigh, more things to look over this weekend.

Yea, these cars are ridiculous, as evidenced by all the no start threads here and on other forums. I never want to hear a bmw guy knock an old Jaguar for being undependable. My '86 jag has never failed to start, and has been many times more dependable than my 88 6er.
Working on a warm no-start now. Was a never start, had a ref sensor go bad while troubleshooting! Fuel lift pump replaced with high pressure pump yesterday and at least it starts and runs well when cold, for about 5 minutes...
Sorry I'm no help, just keep til you get it.

[Brucey]

'no hot start ' is easily diagnosed as being one (or more) of three things;

A- no spark
B- no injector function
C- no fuel pressure

The actions required to address each are obviously going to be completely different to one another and will also vary with the age of the car because the parts fitted are different.

Song needs to

1) - check speed and reference sensor function
2) - check that the +12V feed to the ECU, coil, injectors is good (i.e. present when cranking AND capable of delivering about 10A without a severe voltage drop.)
3) - that there is a 'start signal' to the ECU from the starter motor (also appears on diagnostic connector)
4) - that the coil good
5) - that the ECU is good

However if there is injector function but no spark the priorities above would be different.

I would suggest that Slofut should (if not already done) say what the ABC fault condition is...?

BTW with cars of this age I don't think you can take anything for granted any more. I carry speed and ref sensors in the boot these days, as well as all the tools to change them. I may shortly start to carry a spare ECU too.

Ironically the bits that make 1980's Jaguars go are identical in many cases to those that make 1980s BMWs go; the differences are in how the parts are attached to one another, installed etc and this (together with the usage conditions) determines which parts are liable to fail and how that failure might occur.

Very many of the problems I have seen in cars of this age are just down to lack of basic maintenance, or daft stuff like a failing £2 relay or even just dirty connections in the fuel injection control system.

In these cases diagnosis is the main problem.

You would be amazed at how often there has been one (simple) fault present to start with, but once the diagnosis process is started (usually in a somewhat haphazard fashion), further faults are generated...

cheers

[slofut]

Point taken Brucey.
Of course I felt the need to vent since I've had a tough time sorting this one out. Felt sure I had a fueling problem, and the pre pump was mostly non functional as I think the sock was dirty again, and why I don't know yet as the tank is totally clean. Is it normal for the in-tank sock to be dark or black colored after use? Anyway, I may start another thread or probably follow up on my previous no-start thread as I don't want to hijack. Thanks for your suggestions and patience with frustrated hobby mechanics.

[songzunhuang]

'no hot start ' is easily diagnosed as being one (or more) of three things;

A- no spark
B- no injector function
C- no fuel pressure

The actions required to address each are obviously going to be completely different to one another and will also vary with the age of the car because the parts fitted are different.

Song needs to

1) - check speed and reference sensor function
2) - check that the +12V feed to the ECU, coil, injectors is good (i.e. present when cranking AND capable of delivering about 10A without a severe voltage drop.)
3) - that there is a 'start signal' to the ECU from the starter motor (also appears on diagnostic connector)
4) - that the coil good
5) - that the ECU is good

However if there is injector function but no spark the priorities above would be different.

I would suggest that Slofut should (if not already done) say what the ABC fault condition is...?

BTW with cars of this age I don't think you can take anything for granted any more. I carry speed and ref sensors in the boot these days, as well as all the tools to change them. I may shortly start to carry a spare ECU too.

Ironically the bits that make 1980's Jaguars go are identical in many cases to those that make 1980s BMWs go; the differences are in how the parts are attached to one another, installed etc and this (together with the usage conditions) determines which parts are liable to fail and how that failure might occur.

Very many of the problems I have seen in cars of this age are just down to lack of basic maintenance, or daft stuff like a failing £2 relay or even just dirty connections in the fuel injection control system.

In these cases diagnosis is the main problem.

You would be amazed at how often there has been one (simple) fault present to start with, but once the diagnosis process is started (usually in a somewhat haphazard fashion), further faults are generated...

cheers

I appreciate the tips Brucey, although you have just sent me on a massive research project since I don't know how to do most of the things you listed.
Now I have a few things to keep me busy this weekend, if it doesn't rain.

[Brucey]

re the 'dark sock' issue; I would say this is either rust or something rubber (eg fuel pipe or FPR part) that is disintegrating inside the fuel system. If the crud sticks to a magnet, it isn't rubber.

Of course this cannot cause a 'no spark' condition.

cheers

[songzunhuang]

Brucey,
So I have performed a few tests today and I sent you a PM as well. I didn't get to the ECU part of the diagnostic because I found cause to be concerned with one of the sensors. My understanding is that either of the sensors being bad can cause a no start (or no spark, which is what I have).

Bottom line is that my upper (black) sensor is not providing any reading between pins 1 & 2. This leads me to believe it's bad. I believe this is the speed sensor. The other sensor reads 962 ohms between pins 1 & 2.

Here's what I did check this morning.
1. Main relay is definitely good. Discovered that I could turn the ignition key to the on position, then when remove and replace the main relay, I can hear and feel it "click".
2. Power to the coil was good.
3. Resistance between various coil terminals was good.
4. Fuel pumps buzzing when jumped. Appear to be good.

Just to double check myself, I once again pulled off the coil wire, cranked the engine and looked for a spark. Nothing.

Bottom line, is that I decided to fix the sensor situation before poking around any more. I don't want to make the problem worse.

[Brucey]

Hi Song,
it sounds like you have a duff sensor; the only thing I wondered was if that sensor has a resistance slightly over 1000 ohms, and the DVM is set to 1000 ohm range, the meter may just read 'over range' which often looks a lot like it is reading 'infinity'. If this is a possibility then I'd suggest that you recheck the reading using (say) a range up to 10Kohms or something before condemning the sensor. Testing AC volts from the sensor on cranking is also instructive.

BTW the results you got from the main relay fall into the 'necessary but not sufficient' category. Main relays that are no good at all can give the results you have seen; the relay has a dual contact inside and either or both of the contacts can go bad. Often they go bad in such a way as you seem to have volts, until you draw a load, and then it turns out you don't have volts any more, because there is too much contact resistance inside the relay or something like that.

Needless to say tests that give 'false positives' can be the most misleading of all.

If you want to give the best chance possible of not digging yourself into a hole, make a list of everything you do and keep notes.

BTW I've not seen any PM from you.

cheers

[songzunhuang]

Hi Song,
it sounds like you have a duff sensor; the only thing I wondered was if that sensor has a resistance slightly over 1000 ohms, and the DVM is set to 1000 ohm range, the meter may just read 'over range' which often looks a lot like it is reading 'infinity'. If this is a possibility then I'd suggest that you recheck the reading using (say) a range up to 10Kohms or something before condemning the sensor. Testing AC volts from the sensor on cranking is also instructive.

BTW the results you got from the main relay fall into the 'necessary but not sufficient' category. Main relays that are no good at all can give the results you have seen; the relay has a dual contact inside and either or both of the contacts can go bad. Often they go bad in such a way as you seem to have volts, until you draw a load, and then it turns out you don't have volts any more, because there is too much contact resistance inside the relay or something like that.

Needless to say tests that give 'false positives' can be the most misleading of all.

If you want to give the best chance possible of not digging yourself into a hole, make a list of everything you do and keep notes.

BTW I've not seen any PM from you.

cheers

One thing I should mention is that I replaced the main relay a few months back (October 2015) during one of my other troubleshooting sessions that turned out to be a burnt up rotor. I'd be a bit surprised if the new main relay has failed in side of 5-6 months.

In any case, I have new sensors on the way and in the process of removing the speed sensor the connection at the harness snapped. So I am definitely committed for a new sensor. I hear that these sensors are one of the spare items I should keep in the car anyway.

Just to confirm, a bad speed sensor will result in no spark whatsoever?
For some reason I thought only the reference sensor would cause no spark, whereas the speed sensor would result in erratic running. That's the only problem with a bunch of research. You get a lot of data that isn't always entirely consistent.

[baders]

I think the jury is still out regarding the speed sensor effect. I have seen reports where a dud speed sensor has stopped the car dead, some where it limps. A dud reference sensor and you are stopped cold.

[slofut]

Austin, I remember someone saying that he had a reference sensor that tested normal between the wires and would pulse, but when installed would short one of the wires to ground through the case. Like torquing the mounting bolt would tweek the sensor enough to reveal a potential short from one lead to the case, causing a no-start.
How is the car coming along?

[Ralph in Socal]

Austin, I remember someone saying that he had a reference sensor that tested normal between the wires and would pulse, but when installed would short one of the wires to ground through the case. Like torquing the mounting bolt would tweek the sensor enough to reveal a potential short from one lead to the case, causing a no-start.
How is the car coming along?

One way to test that would be to try and start the car without bolting the sensor in place. The loose interference fit of the sensor is enough to keep it in place for a short time while trying to diagnose.


Ralph

[songzunhuang]

Wanted to close the loop on my no spark issue.
The bad speed sensor was in fact the problem. Ralph sent along a replacement and I put it in this evening. The car fired right up.

I can say with absolute certainty that my car will not run with a faulty speed sensor.

I also recorded this event in my project thread. I just wanted to close the loop in the various places that I posted.
Thanks everyone for the tips.

[Brucey]

glad you got it fixed.... (item #1 on my list.... I guess I just got lucky...? )

now, I wonder how AustinK is getting on....?

cheers

[Chris Wright]

I might as well throw my 2 cents in as well. (That way the sensor cables will fix the problem and this post won't be needed!!)
____________________________
The power flows from the ignition switch, through the immobilizer (OBC Relay) and energizes the Main relay. The output of the Main relay in-turn energizes the fuel pump relay. To check the immobilizer, go to the heading: Checking the OBC Relay: below.
_____________________________

This procedure "No Start" procedure is for cars from '83 to '87, the '88/'89 have their main relay wired slightly different and different Ref. and Speed Sensors.

Break down of components needed to get spark to the engine:
Ignition Switch, OBC Relay, Main Relay, Coil, Reference Sensor, ECU
____________
The Engine Cranks Over: First check for spark:

Pull a plug wire and get an extra spark plug* and plug it in. Position the plug so that the metal shell is grounded and try to start the car. Watch the plug to see if you are getting a spark. If you are getting spark, skip to the next section to check for fuel.

If you are not getting spark at the plug, test again at the coil tower. if you get spark at the coil but not the spark plug, check the Distributor Cap and Rotor.

(* or use a spark tester, these can be had from an auto parts store. You may also need an extra plug wire to plug in and test at the coil tower.)

If no spark from either the Coil or Spark Plug:

Checking the ignition Switch and Coil:
Check, with the key on, for voltage at terminal 15 (+) on the coil (voltage will appear at Term 1 also). If no voltage, the ignition switch or wires to it may be bad. Wiggle the ignition switch to see if it is intermittent**.

Check, with the key off, the resistance between Terminals 1 (-) and 15 (+), of the coil. It should be around 0.50 Ohms. The resistance between terminal 15 (+) and the high tension center tower should be around 6000 Ohms. Replace the coil if it has high or infinite readings (Open) or little or no resistance (Short).Checking the OBC Relay:
Next, pull the Main relay and with the key on, check for voltage at pin 86 and pin 30 in the relay holder (the numbers are molded into the bottom of the relay).

If there is no voltage at pin 30, check the Red wires at the Positive Battery terminal (Clean and tighten them regardless).

If there is no voltage at pin 86, check the OBC relay box under the driver's side kick panel, by jumping:

New Style OBC (10/'84):
the two green wires (pins 3 & 4) in connector C2.

Old Style OBC:
pins 14 (GN) and 15 (GN/RD) in the black connector C1 (See page 116/117 in the ETM for pin locations).If there is no voltage at pin 3 of C2 (New Style), or pin 14 of C1 (Old Style), the ignition switch may be bad (both terminal 15 on the coil and the OBC pins above are powered from pin 15 of the ignition switch).

If voltage is present at pins 86 and 30: It is easiest to just jumper the Main Relay as in the Photo below or swap in a new/good main relay and then skip to the "Checking the Speed and Reference Sensors" below. Otherwise check the main relay as in the "Checking the Main Relay" steps after the Photo.
Jumping the Main and Fuel Pump Relays:
Image
_________

Checking the Main Relay:
Check the circuit diagram on the side of the Main Relay to see if it has two separate contacts for the output pins 87 and 87b or if it just has one contact going to two pin 87's. Clean the relay prongs and the sockets in the holder base and plug the Main relay back in.

If there is just the one contact: With the key on and the Main relay plugged back in, check for voltage at the RD/WH wire in an fuel injector plug (or use a "noid" light on the injector plug).

If there are two contacts: With the key on, check for voltage at BOTH terminals 87 and 87b of the Main relay. (You have to wrap a wire around each pin and plug the relay back in and test for voltage at each wire. Don't short between the pins!! One terminal can be good, suppling power to the fuel pump, and the other bad, not suppling power to the ECU, or vice versa.

Alternately, to check both relay contacts and the Start Input: With the key on, check for voltage at Pin 4 (Start Input) and Pin 35 in the ECU harness plug, and at the RD/WH wire in an fuel injector plug (or use a "noid" light on the injector plug). If there is no voltage in the tests above: replace the Main Relay. As stated, it might be easier to just replace the Main relay out of hand in the first place, instead of doing the testing. It is also a good idea to keep a spare Main relay in the trunk Tool Kit and just replace it (and the Fuel Pump Relay) out of hand if the car stops out on the road. Checking the Speed and Reference Sensors:
Check the plugs to the sensors first, the plastic can get brittle with age and they can crack around the wire locking bail leaving them loose. Then check their resistance as below (won't hurt to check them both hot and cold if the car's hard starting is temperature sensitive):
<a><img src="http://bigcoupe.com/ims/pic.php?u=21GvpX2&i=349" border="0"></a>
_______________

The resistance test above is simpler, and gives you info on the condition of the crank sensors, but the AC voltage output test below is a bit more conclusive, as it also checks for the rare instance of a missing Ref. pin for the Ref. Sensor in the flywheel (Automatic transmission only):

1. Pull rubber boots back off of the two plugs on the sensor lead side (not the harness side). Pull the main relay & fuel pump relay to disable the FI system.
2. Set your voltmeter to AC volts and back probe the black and yellow wires while cranking the engine.
Speed Sensor (Black Connector - reads flywheel Teeth) -- 4.0 AC volts max while cranking the engine.
Ref. Sensor (Grey Connector - reads flywheel Ref. Pin) ------ 0.4 AC volts max while cranking the engine.
_______________
__________________________
Chris Wright
'87 635Csi
3 MGA's
MG 1100