The ends of an HT coil go to two plugs, so two coils serve four plugs. Only three lower voltage wires needed.
The spark goes to two plugs simultaneously; one on compression stroke and the other on the valves open stroke, which makes timing simpler. BUT if there is an open-circuit plug then maybe no spark on the other one, and it leaks away elsewhere (blue sparks might be seen outside the leads in the dark) or it breaks down (cooks) the coil insulation.
A lot of cars did this trick in the 90s, some merely switched the 12V to the coil in the traditional manner, others used a CD system so you might have a 400V pulse (ouch) coming in to fire the coil.
It looks like the NA feeds 12V (Blue wire) from the ignition switch to both coils, and the Igniter block (some power transistors) switches the earthy end of the respective primaries (Red and White wires) to ground.
Richard - thanks for that, I was unaware of this type of firing order. In fact, now you mention it, I seem to recall the expression ‘wasted spark’ from the past, so that now makes sense to me. However, this means one coil is working twice as hard as a four coil system, so no wonder some get coil problems.
A double check on my wiring diagram shows a live 12v feed to one pin, and a switched feed from the PCM on the other pin, with ground on the centre pin between the other two. So it seems I too have a wasted spark system. There you go! - We still learn something every day.
So - I can only assume the NA coil pack is grounded via the mounting bolt(s), if mine has a common ground between both harness plugs. (According to Robbies description?)
Thanks for all your comments guys your suggestions are going to give me plenty to work though as Robbie said there is only one live supplying the two coils so I think I’ll start there and work back checkingbthe wiring and earth cables I still think it could be the igniter and as I xan pick one up for a tenner mite be worth a gamble changing it oout. Its just frustrating because we have had it running when we first picked it up a few days ago
Ben - what are you calling the ‘ignitor’? - The coil provides the high voltage, and the spark plug lights up. Both at the command from the PCM (ECU) which in turn is part controlled by the CAS and I assume the Crank sensor? My main suspicion is a break in the 2 and 3 wire from the ECU to the Coil, which is one wire out of the three I now know are in the harness plug.
To be sure, at this point, briefly spray some WD40 into the end of the harness plug and the corresponding Coil Pack Socket, then part push the plug into the socket, but not so it clicks in. Move it back and forward slightly and then push it firmly in place. Try starting now - if it doesn’t work, then it’s the next step -
You need to disconnect the harness plug from the coil pack, read which is the steady 12v supply (which pin) and then put the meter lead (if it has a sharp point on the end) into one of the other two pins, Stick the other lead on ground (Chassis) and get someone to crank the engine for a few seconds. You will read a voltage on a good pin, if so, repeat the process on the other pin. One of these two won’t read, if I’m right. You will get a pulse - meter needle will flip - which indicates the pulse from the PCM to cause the spark cycle, no pulse - no signal = duff connection.
If you can establish this, then if both pins show a pulse, then the fault is elsewhere, so could be the CAS, but you need to check the wiring out before you start changing other components.
If there’s no pulse on one wire, then it could be one of two things, either the wire has pulled out of the pin in the plug, or it’s either broken or making a bad connection elsewhere. After this point I can’t help you further, as I have no idea where breaks can occur, or what connections are at the ECU - or the harness plug end that goes from the engine to the ECU.
The early single-moulding coilpack with only a single three-pin connector and four sparks does not need a ground, because this has the external igniter.
The later NA coilpack with two mouldings and two LT connectors has the igniter built-in, so that version has a pair of three/four-pin connectors (one for each coil of two sparks, 1&4, 2&3). The wires are +12V (Blue), feedback for diagnostic checks and tacho etc (Black/White), ground (Black), ignition pulse from ECU (Brown for 2&3, Brown/White for 1&4). Some packs have three pins fitted (ie leaving a gap) some four, so maybe they drop the tacho/feedback line.
The pin sequence on each of these later connectors (L-R, latch at the bottom) is
O/K - I get it, weird and complicated. The term ‘Ignitor’ threw me, as I associate that with earlier diesels that needed a glow plug to start. (IIRC, in the states they call that an Ignitor.)
The ‘Igniter’ is just a transistorised switch instead of the old fashioned points. There will also be some other bits in there intimately associated with it to protect the transistors and ECU (and radio reception).
I was thinking about this afterwards: if the tacho feed is active and being used, then a missing spark will show up as only half (or even no) revs on the tacho, for instant diagnosis by the driver of an electrical fault (as opposed to fuelling problems).
A very frustrating afternoon spent in the garage and still no idea why it won’t start. I think the problem may be my lack of understanding of automotive electrics and only very basic knowledge of using a multimeter. The carpet over the ecu is wet so will remove the ecu and dry it out. Just in case.
Think it may be time for a fresh pair of eyes but don’t really want the cost of a auto electrician.
This is a known problem, a wet ECU can cause all sorts of odd issues. Shame you didn’t find it before. Hopefully if you remove and dry the ECU, clean all the connectors to it, you could be OK.
Hindsight is a great educator, but there are many threads on this and other mx5 forums about problems caused by water getting into the ecu from blocked roof drains.
Next thing to do is check your drain tubes to stop any more water getting in.
Good luck.
Combermartian thanks for that it’s not a problem with the drain tubes this project has been stored in a garage with the roof down but let’s say the garage roof wasn’t water tight
That seems a likely culprit then, and solves the puzzle of why it started at first but wouldn’t do so later: If it was sitting with a dead flat battery when the ECU got damp then it wouldn’t have much effect beyond a bit of surface corrosion, which would explain why you were able to get the car to start straight after fitting a new battery. But since then, there’s probably been a bunch of little electrochemistry experiments fizzing away around the damp ECU pins.
I should unhook the battery straight away, and then get the ECU out, clean it up and dry it out. Might well be fine after that.
OK quick update still not starting turns over quickly seems to fire but won’t catch and run.
Parts changed so far for new or tested parts.
Fuel filter fresh petrol
Spark plugs
Igniter
Coil pack
Ht leads
Air filter
Removed ecu and checked for water damage but all looks clean with no damage I can see. left it in the house in the warm for 36 hours just to make sure no damp present
We know fuel is reaching the injectors and all four are firing . But still only spark to 1 and 4 I have ordered a diagnosis code book and will run a diagnosis check next but if that doesn’t show up any problems I think the next step is to change the cam angle sensor.
I was wondering about fuse / relays but with it giving spark to one side of the coil I think we can rule that out as there is only one power feed to the coil pack
If you have a multimeter, then it is time to look at the three leads going to the coilpack from the igniter. (Don’t use a bulb)
Set the multimeter on AC Volts and a higher voltage range (just in case, it can be reduced later for more sensitivity).
Disable the fuel pump so as not to put petrol into the system while testing.
Unplug the lead from the coil pack.
Connect the AC voltmeter across Blue and Red at the unplugged end of the lead, spin the engine briefly and note the voltage.
Connect the AC voltmeter across Blue and White at the unplugged end of the lead, spin the engine briefly and note the voltage.
These voltages should be the same within a volt or so.
If not then the Igniter, or the lead to the coilpack, or the lead from the ECU, or the ECU is faulty.
Checking this again at the Igniter end will eliminate the coilpack lead.
Checking the input wires Brown and Brown/Yellow relative to Black (ground) will eliminate the igniter.
Checking again on the wires Brown and Brown/Yellow relative to Black (ground) at the output of the ECU will eliminate the leads between ECU and Igniter.
If the voltages matched on the initial test then plug the lead back into the coilpack
Repeat the initial test with the AC voltmeter but now looking at the terminals on the igniter end of the lead.
If the two AC voltages are different from each other this time then there is a problem with the coilpack or the connectors.
Followed Richards suggestion and have found badly connected wires going to the cam angle sensor and another bad connection to the coil pack. I’ll try to upload a photo my question is how to solve the problem.
Injector relay? You should here the fuel rail pressurising when ignition is turned on.It too can be bridged but I forget the pins right now.Or try small spray of “Easy atart” or similar directly in to inlet manifold.Beware do not use too much .Just light mist for about 2 seconds and see if cars fires up.I usually try this while cranking the car over.Thsi will eliminate electrical issues etc