10.9V is near the ECU's 'wake up and get to work' threshold (generally asserted to be ~ 10.3V to 10.5V). Our Spider (1984) reads 12.6V with everything off and ~ 11.8V during cranking. I'd suggest a careful investigation for voltage drops - iffy connections, corroded terminals, loose crimps, etc. A tenth of a volt here and a tenth of a volt there could add up and conspire against you. And some might be intermittant such that only when they are all acting up does the car fail to start.

My brother is into old jeeps but he has a page about electrical stuff that is universal enough to apply to our cars. It includes info about 'voltage drop testing'. (and if you have some time to waste, browse around his 'marriage & jeeps' section - some of that will apply to Alfa's, too.)

 

It may be the ECU, if you need a "check spare" I have one.

I bought one on eBay and I suggest you do the same. They run average $150 or so sometimes more sometimes less.

The most "friendly" scenario is for you to send it to someone to try in their cars. Most people would be leery to send out an ECU for trial in a car at "arms length".

I note that Vick seems to have a swap service, you may look into that. There are other ECU places that advertise "repair", but I would be very scheptical.

My thoughys, Elio

 

Thanks for the help, I will will give the Jeep page a look, I do have an 89 Grand Wagoneer, so it will be good info. I have done quite a bit of ground and eletrical cleaning, but I can always do more. Another thing that makes me think it might not be the volt drop is the fact that sometimes when this happens, I can crank on it for a while and it will start up, even after about 30 seconds of cranking. It really acts like a relay more than anything.

I will give Vicks a call on Monday, price the rebuild service, I think that maybe the way to go if infact it is the ecu...

 

Can anyone tell me where the ignition ECU gets the volts reading from? I think that pin #4 on the ignition ecu gets the volts reading, but from where? Is it the fuse box? I read on an old Alfa Digest that it comes from the starter solenoid, any truth to that?

 

Send esteemed BB member PapaJam a PM, include your email address and ask for copies of his excellent color-coded wire diagrams. He will reply with some attachments that will make figuring out your Spider's electrical issues much easier.

Off hand, I think the ECU gets its electrons via a green/black wire off the fuse box. But, since that tidbit is via memory and I have CRS disease, verify that before proceeding further.

 

Below is a snip from PapaJam's wire diagram for the ignition ECU (his chassis diagrams are all color coded). And also a snip from the Alfa Shop manual re: ignition ECU testing. Note the tests described in the shop manual are not directly to the ECU but rather on the female sockets in the wire harness - disconnected from the ECU. Do not use an ohm meter on the pins on the ECU itself. AFAIK, there is no practical test of the ECU itself. If all sensors and inputs test good, the only thing left is a faulty ECU. But before you rush to condemn it, make sure everything else is 100%.

 

Sweet!!! Thank you very much, any idea where the violet wire attaches to the ignition switch? I had problems with my starter, even after a new starter and switch it would still sometimes not start, so I installed a relay, I wonder of the same "not enough juice" to start the starter is keeping the ecu from firing....

 

Again, courtesy PapaJam's wire diagrams (edited for specificity) and the Alfa Shop Manual:

 

SamW:
I am experiencing the same problem. I have redone all of my grounds, etc., etc.

How do you test a relay? I need to test the two next to my ECU. I bought a new main relay, but that was not the problem, so now I have two of them. And, how do you test the flywheel sensors and the vacuum sensors. If you would prefer not to take up space with your reply, then PM me at mandlebaum at sbcglobal dot net.

Many, many thanks
Dean

 

Eric,

OK, I guess my question is how do I get from the black wire on the switch to the starter solenoid to the voilet wire going to the ecu? They have to connect somehow, am I thinking about this right? Does the voilet wire splice into the black wire?

Thanks for the help!

Dean,

If you have a couple of wires, unplug the simple relay and get a spare battery (I keep a little motorcycle battery for stuff like this) hook the positive side of the battery to pin 85 (it should be marked on the relay) and the negative side to pin 86, then check the resistence between 30 and 87, should be no resistence when 85 and 86 are hooked up to the battery, if you unplug one from the battery, 87 and 30 should be an open line. As for the vacuum, I just listend for this one, you should hear it click when you start spinning over the engine and then click again after the engine stops spinning with the starter. I did not check to make sure it worked, I assumed it was working because of the click. For the flywheel sensors go to hiperformacestore.com and look under the Spider L-jet, he gives a really good description, you test the resistence between two of the three pins, but you will need his photo and range to get it right. I don't mind sharing the post, we are chasing the same thing!

 

...how do I get from the black wire on the switch to the starter solenoid to the voilet wire going to the ecu? They have to connect somehow, am I thinking about this right? Does the voilet wire splice into the black wire?

Yes, up under the dash next to the stering column you should find where a short cable exits the column surround. It has two connectors, one with two wires (a third spade connector is not used), the other with one wire. The violet wire splices in where the black wire connects to the short harness from the ignition switch.

If your model Spider has the knee bolsters they may be somewhat in the way. But that is where the purple & black connect.

 

A relay is essentially an automatic switch. A 'trigger' signal energizes a small electromagnet which then makes or breaks internal contacts to 'relay' the signal to make something happen. The other relay (called the drive relay) in your Spider is a little odd in that its trigger is a tachymetric signal (pulsing) from the coil. The tachymetric signal tells the drive relay the engine is running (as the coil charges & discharges to make sparks) and the output from the drive relay powers on the fuel pumps. If the drive relay does NOT see a tachymetric signal (engine not running) it turns off the fuel pump's power supply.

So, to test a relay you need to check for input (the trigger signal) and output (what ever the relay is supposed to make happen).

To test the flywheel sensors, peruse the Spider L-jet diagnosis page. Link is in my signature.

The most common problem with the vacuum sensor (aka MAP sensor) is that it leaks (will not hold a vacuum). Does your's moan, groan or whistle? If so it is likely leaking. BTW, be sure you are sitting down when you check the price of a new vacuum sensor...

 

Ahh!!! That is what I was looking for, someone (PO) installed a serious stereo and they spliced wires all under the dash, I have been slowly putting them back to normal as I find them, I replaced the switch, so I should be able to find this, I was not sure if I needed to start looking at the fuse box or under the hood. Thanks again!! After work, all these test will be checked and I will let you now what I find!!!

 

Sam,
The diagram in post #9 has an error in that the violet wire should connect to the ignition switch connector in location #1 (wire #50), NOT #2 (wire 15/54). The diagram has since been revised to correct this error.

If the violet start signal wire on your 89 is like the 86-88 Spiders, this wire does not go directly to the ECU. It first makes a stop at terminal 87b of relay N10 (interior light timer) then makes it's way to connector G121 located behind the glovebox where it then splits 5 ways; one each to the ignition & fuel ECUs, one to the Thermo Time Switch (TTS), one to the Cold Start Injector (CSI) and the fifth goes to terminal #50 of the drive relay (to energize the fuel pumps during cranking). Since this is from memory, I'll check the diagrams this evening to be certain.
If you'd like the diagrams for an 88 Spider (if not seen Alfa published diagrams for an 89), please PM me your email address.

 

Thanks Jim, corrections noted! I sent you a PM, but I will keep the thread going for the benfits of others who might follow.

I think my '89 is an early one, I have the L-Jet fuel system. Is the N10 relay in the fuse box? My interior lights don't turn on unless you do it manually, so it is worth looking at that too. The G121 connector, when you say it is behind the glove box, can I get to from under the dash, or am I going to have to pull the glove box out again like I did when the cowl drain hose split?

 

Is the N10 relay in the fuse box?

Yes.

My interior lights don't turn on unless you do it manually, so it is worth looking at that too.

Before getting all happy about trying to find the all but unobtanium, and horrifically expensive if you do find one, you might wanna check/clean the contacts at the doors and make sure that the lamp housings are actually wired correctly as the lamps can indeed be mis-wired and still work as you describe. (door switches are just grounds, or rather contacts that engage ground when the door(s) are opened)

 

OK, I went through all the test and everything came back A-OK, so I guess I have a dead ignition ECU... Anybody got a Bosch 0-227-400-003 for sale?

 

Give APE a call:

APEDIRECT.COM

209-833-8330

 

Anybody have a picture of a series 3 fuse box? I looked at mine and I have a normal silver realy where the linterior lights timer is supposed to be, I am guessing that the longer black relay next to it is the interior light timer, but given that one is out of place, I want to make sure the others are correct... (D@MN POs!!!!)

 

Early or late Series 3 Spider fuse box? First sketch is early S3 ('bullet fuses'), the second is the late S3 (blade fuses).

Sketches are from either the owner's manual or shop manual. The photo is the fuse box in our '84 Spider. I bypassed the light timer relay when it died.

 

Cool! Late series 3, my quad is an '89. Do you remember what the interior light timer looked like? I had a normal silver relay plugged into the N10 spot, like the one in your picture plugged into I12. What was plugged into I12 looks like what you have in N25, and what I have in N25 is a red version of what you have in N25... Plus I have another item in I2, which makes sense since I have the hardtop with the rearwindow heater. Hmmm.... I also have a plain silver relay in the acoustic warning spot (I3), is the buzzer built into this relay????

I think my fuse box is FUBAR... (Things work, but like the one way valve for the vent of the fuel tank, just plugged in wrong....)

 

I also have a plain silver relay in the acoustic warning spot (I3), is the buzzer built into this relay?

No, the buzzers are located on the steering column (2 gray or white boxes)

BTW, 'acoustic warning' in the book and diagrams means 'horns', not buzzers.

 

The light timer relay in our '84 (early fuse box) was black plastic. I have a couple of dead ones if you want 'em...

 

Do they have some markings on them that might help me identify them? Is the buzzer built into the top left relay?

I have a bunch of relays and stuff from Milanos, maybe they are interchangable....

 

OK guys, the mystery deepens... I got the new ecu, plugged it in and nothing changed, still no start. I checked the signal to the (-) side of the coil and it was getting nothing (had a thought maybe my coil wire was bad) so the coil was getting nothing from the ecu. Went through all the tests again and everything passed with flying colors, so I plugged the ecu back in (all the measurements are at the terminal for the ecu, so you can find a bad wire) tried again, nothing, tapped on the ecu a few times, nothing. Then, just for kicks, I plugged my multi meter into slot #87b of relay N10 in the fuse box just to see if I was getting the same 10.9 volts there that I was getting at the ecu, well I did not get to see what the volts were, because the Spider started right up... Why? would there be a bad connection at the fuse box that might get the voltage to drop at the ecu, but the test said to check for between 6 and 10.3 volts, I was getting more, so I thought that was good. I need to check the volts at the ecu again...

 

Sam:
First let me preface the following by saying I do not pretend to be an expert on Bosch wiring systems. By comparison the wiring on my old British iron was elegant in its simplicity. But juice flows from positive to ground just the same( or from negative to positive prior to 1967 that is). How it completes the circuit is the confusing part.

Do you have one of Papajam's wiring diagrams? Looking at mine there is power coming to the (+) side of the coil, on the green & black wire from fuse#7. The (-) side of the coil goes to ground through the connection to the ignition ECU. If I understand it correctly, current if flowing from fuse #7 to the coil and to ground via the ECU. The ECU controls the grounding and hence the function of the circuit. I would not expect to see power at (-) on the coil if there was none coming from the (+) side.

That the car started when you jiggled 87b at the light timer is I think a bigger clue. The wiring diagram shows battery power going direct to the fuel injection ECU and thence to the ignition ECU. A violet wire runs from the ignition ECU to the black wire that goes from the ignition switch to the starter via a shared junction at the said 87b, which is identified on the wiring ignition system sub-diagram as the start signal. I would look behind the fuse box at this junction point and clean/test this connection.

 

Andy,

Thanks! On testing the coil, I hooked up my multi meter with the rpm/tach feature, it hooks to the (-) side of the coil, the same as the tach signal, it measures the signal from the ecu, that is how I tested it and found that no signal was going to the coil. I do have Papajam's wiring diagram, and I have to admit it is one of the easiest ones I have seen to read.

I agree, I am about 90% sure that the problem is in the fuse box, my issue is that if I was reading enough volts at the ecu, why would it matter what is in the fuse box? the two voilet wires are connected together with a single female connector that goes to N10. I don't see an issue with it, I will redo the connection just to be sure, but the green wires to the same N10 connector just above has a melted look to it, but these just go to the interior lights. The car did not want to start again, so I dropped the fuse box and played with the wires, now it starts again. I will spend some quality time cleaning my fuse box and see if that clears up the issue...

 

I don't see why you couldn't install a jumper between the violet wires as a test to bypass a potentially bad junction. Power only moves through the circuit when the key is turned to the "start" position, and you need that that bring the ECU up and get the cold start going in the fuel injection system. I suppose they made a junction at the interior light timer to shut it down when starting?