Swole Ignition coils, blowing fuse and big problems on my Jazz

[CRC]

Have a look on YouTube and search for coil pack testing. There are lots of hits of varying quality, but the gist seems to be that you measure across the terminals with a multimeter  set to 20 k ohms and this enables you to measure the resistance (probably impedance) of the primary and secondary coils.

It would be good to do this on the 8 old presumably Honda coil packs and compare the readings. If they are all very similar to each other, it's possible that they are all still ok, however, marked differences would suggest the opposite.

Then, obviously, do the same on the 8 new coil packs.

Looking at Porsche Boxster forums, it seems that there are a lot of variables between manufacturers.

Compare the results and look for any obvious differences that might point the way to that Eureka!! moment...

[CRC]

Regarding following those two wires any further, they will have to come through the bulkhead to get to the interior fuse box, but as you can't find any sign of a connection to earth, it's unlikely that you'll find anything.

A while ago, I bought a cheap 30 amp moving coil ammeter to put in series with a wire I was having issues with. Cost about £7 from eBay but did show the problem.

It would be interesting to put this in the line to show the amps being pulled.

Multimeters mostly have amp measuring settings but are normally fused at 10 amps .... and most fuses have been blown by people measuring voltage when on the amps setting ...which causes a short circuit through the multimeter.

My job before I retired was designing control systems and writing the software for PLC controllers for process departments in major food factories, and getting the wiring right in the first place was key to being able to fault find the software later on. Testing wiring is never wasted time ...

[CRC]

For info, I've just checked on Autodoc and there are dozens of coil pack manufacturers there, with the prices ranging from £20 to £60 each.

Don't despair just yet, but I would say they are definitely Chinese ... but could be very good ... who knows?

[Dwight]

Thanks for those informations. I'll definitely check the coils tomorrow.

I was trying out chatgpt, see if it could find any result in japanese search engines maybe, since nothing comes out in english for me.

Anyway, by talking more about the problem and giving more details, it came out with this interesting conversation:

Chat GPT - Let’s restate the pattern:

✅ Starts fine

✅ Runs fine

❌ Fuse #24 blows while idling

❌ Not during cranking

❌ Not immediately key-on

🔁 Happened while parked, engine idling

That rules out a lot.

🔎 What This Tells Us

If it:

Doesn’t blow at key ON

Doesn’t blow during crank

Only blows while engine is already running

Then the short/overcurrent is dynamic, not static.

Meaning:

It is triggered by something that only becomes active once the engine is operating.

That dramatically narrows the suspects.

🎯 Top 3 Most Likely Causes Now
🥇 1. Alternator Internal Failure (Very Strong Candidate)

At idle, once the engine is running, the alternator is:

Producing current

Charging system at ~13.8–14.4V

Under load if fans, lights, etc. kick in

If a diode inside the alternator is failing:

It can create voltage spikes

It can create internal short events

It can backfeed into ignition circuits

It can momentarily overcurrent a branch

This does NOT happen during crank.
It happens only once running.

And it can happen randomly at idle.

This matches your symptom perfectly.

🥈 2. Heat-Triggered Short in Harness

At idle:

Engine warms up

Exhaust manifold heats nearby wiring

Plastic insulation softens

If the front bank harness runs near:

Exhaust manifold

Cylinder head

EGR pipe

Then once hot, insulation may soften and allow intermittent grounding.

Cold → fine.
Warm idle → pop.

Very common on older Hondas.

🥉 3. Failing Ignition Relay (Main Relay)

Honda main relays from that era are famous for:

Internal solder cracks

Heat-related intermittent faults

At idle when cabin warms up, relay heats, internal contact resistance increases → surge → fuse blows.

Less common, but possible.

🚫 What This Is Probably NOT

Not a bad coil (they're new, and failure would usually be immediate)

Not a constant hard short (it would blow instantly every time)

Not a starting circuit issue

This is a heat / charging / dynamic event.

🔬 The Single Best Test You Can Do Next
Test the alternator.

Very easy:

Replace fuse.

Start engine.

Disconnect alternator main connector (small plug, not the big battery cable).

Let engine idle.

If fuse never blows again → alternator is the culprit.

If it still blows → not alternator.

⚡ Also Check This

While idling, measure voltage at battery:

Should be 13.8 – 14.4 V

If you see spikes over 15V → regulator issue

If voltage fluctuates wildly → diode problem

🔥 Why I Lean Toward Alternator

Because:

Happens only while running

Happens at idle

Random

Front bank only (could be how that branch is routed)

Alternators with bad diodes create weird, non-obvious fuse failures.

And many people chase wiring for weeks when it’s actually the alternator.

One More Important Question

When the fuse blows at idle:

Does the engine die instantly?

Or does it keep running but rough?

Do warning lights come on first?

That final detail will pinpoint this even further.

_______________

Me - To answer your question, when the fuse blows at idle, the engine runs only on bank 2 (rear) of ignition coil. Weirdly enough, trying to put back the fuse 24 and it blows immediately this time, as if something had been triggered now and you can't replace the fuse because it will keep on blowing. So yes, the engine runs solely on still functioning rear bank of coils, which are running on fuse 2 as said before. But since it's an engine designed to run with 8 coils, running on only four make it run badly. Kind of as if I was on limp mode.

Eventually, fuse 2 pops too since when this issue started for the first time, fuse 24 blew, then a few minute lates at idle fuse 2 blew too.

I think that maybe fuse 2 blow if fuse 24 blows because the engine is not supposed to run only on those 4 coils. But I don't know. Not sure. Maybe it's the same thing blowing fuse 24 first then blowing Fuse 2 eventually.

Also the engine light flashes when any of those fuse blows and the engine start a rough idle for the records.

_______________

ChatGPT - Ah — now the picture is crystal clear. This is actually very valuable diagnostic info. Let me summarize carefully:

🔹 Observed Behavior

Engine starts and runs normally.

Fuse #24 blows first while idling.

Engine then runs only on rear bank (#2) → rough idle, limp-mode-like behavior.

Attempting to replace fuse #24 immediately causes instant blow.

After some time, fuse #2 may blow too, leaving the engine completely dead or in severe limp mode.

Engine light flashes when fuses blow.

Symptoms only occur while idling, engine running, not during start.

Wiring visually looks fine; coils are new.

🔎 Key Takeaways From This Behavior
1️⃣ Fuse #24 Blowing First Is the Trigger

Front bank (fuse #24) blowing first is the initiating event.

The engine still runs on the rear bank (#2) because those coils are still powered.

Running on only one bank of 4 coils is not normal, so the engine ECU may try to compensate by limiting timing and fuel → limp mode behavior.

2️⃣ Fuse #2 Blows as a Secondary Effect

After the front bank is lost, the ECU may try to run everything with remaining coils or injectors, overloading the rear bank supply → fuse #2 eventually blows.

So fuse #2 blowing can be a consequence of fuse #24 failure, not the root cause.

3️⃣ This is a Dynamic / Overcurrent Fault

Instant blow of #24 after previous trip indicates something is already in a faulted state when power is applied.

Not a random fuse: something in the front bank ignition circuit is already shorted or drawing excessive current.

4️⃣ ECU / Alternator / Power Supply Suspected

Since wiring and coils appear fine, and the engine is running at idle when the fuse blows, likely causes include:

ECU internal fault on the front bank driver circuit

The ECU switches the front bank ground side of coils.

If a transistor inside the ECU is failing (shorted), it will draw full current from fuse #24 immediately, blowing it.

This explains why fuse blows instantly when replaced while the fault persists.

Front bank shared wiring issue under load

Less likely since harness looks good and you can see no wear.

But still possible: intermittent internal short in connector near ECU or coil plug.

Alternator / voltage spike problem

Alternator faults usually affect the whole ignition system; but your rear bank survives for a while, making this less likely.

🔧 Why the Engine Runs on Bank 2

ECU design: if one bank loses power, ECU keeps firing the other bank to avoid engine stall → rough idle / limp mode.

If front bank short pulls high current → rear bank may eventually be affected by shared grounds or voltage sag → fuse #2 eventually pops.

This matches the fact that rear bank fuse blows later, not immediately.

🔥 Most Likely Cause Right Now

Given:

Instant blow on #24 when replaced

Front bank isolated problem

Coils are new

Wiring visually good

The top suspect is the ECU itself — the front bank driver is failing internally.

These are usually low-side transistor switches inside the ECU that control the coil grounds.

If one fails shorted → fuse #24 sees full 12V → blows immediately.

ECU replacement or repair of the ignition driver circuit is often required.

✅ Suggested Next Diagnostic Step

Do not just replace the fuse repeatedly.

Instead:

Disconnect front bank coils completely (unplug all 4).

Then insert fuse #24.

If fuse now holds → fault is downstream of the fuse (likely ECU driver or coil connection).

If fuse still blows → fault is before coils, e.g., ECU output driver is internally shorted.

If disconnecting coils prevents fuse blow, check each coil connector and ECU pin for shorts.

If fuse still blows even with coils unplugged → likely ECU ignition driver module is defective.

💡 Important Observation:

This explains why:

Fuse blows only while running (driver transistors are active, idle running keeps them energized).

Fuse blows instantly when you try to replace it after initial event (driver is shorted internally).

Rear bank survives briefly but may be overloaded afterward → fuse #2 eventually blows

It's not a short read, but it's interesting. What do you think about that alternator and/or ECU theory?

[CRC]

I'm looking into some of your questions, but in the meantime, as you seem to be the only person in the world who has had this problem (it's normally me!!) I'm looking for something obvious that we might have missed.

The attached files ask a couple of questions ..... what is that strange looking grey wire(?) shown in loom1.pmg and is it me, or does that connector look badly damaged in the loom2.png?

Looking forward to your coil pack resistance measurements .... if they look about the same, you almost have to conclude that the problem isn't being caused by the coil packs.

[Dwight]

Yes, the fact nobody seem to ever had this issue is absolutely both exciting and demoralizing, lol.

You've got good eyes, but on your first picture, I'm pretty sure that's a little wire where previously was still attached a paper tag stating when was the last service, I have a few of those hanging in the worshop and it usually hold better throughout time than a sticker... I'll still check it out to make sure, but it looks like this is it.

About that second picture, it is just a part of the big squared plastic case arround the harness that suffered a bit because of me, my Neandertal delicacy and a screwdiver to open it... 

Did you take a look to the conversation with ChatGPT?

I know it's flawed and I'm not really a fan of A.I in general, I avoid it as best as someone can in 2026, but in this matter, I thought the conversation was still interesting. I don't know if it's correct in its statement, but when redoing the conversation from the start, giving him as much of the story as I gave here through this topic, it still ended up firmly believing that my ECU was bad.

I quote:

ECU ignition driver (front bank)
This is the most probable culprit, and here’s why:
•   Front bank fuse (#24) blows first, only the front bank loses power
•   Once fuse blows, replacing it instantly causes it to blow again → short exists on front bank circuit
•   Rear bank is fine initially → indicates bank-specific fault
•   Dynamic / heat related → ECU driver transistors (low-side switching) are activated only when engine is running
•   P0351 and P0357 codes (coil A & G) directly indicate primary/secondary coil circuit malfunction → usually points to ignition driver / wiring issue
•   The fault progresses over time → could have slowly destroyed coils
How it works:
•   Each coil is switched on the low side by the ECU via a transistor
•   If transistor fails shorted → full current flows through fuse → blows
•   Once fuse blows, front bank is dead → engine runs on rear bank only
•   ECU may try to compensate → puts extra load on rear bank → fuse #2 eventually blows
This matches all symptoms perfectly.
________________________________________
5️⃣ Alternator / charging system
•   Alternator faults usually affect the whole system, not just one bank
•   Rear bank survives → fuse #2 blows later, consistent with secondary effect
•   Alternator is less likely as primary cause, but could exacerbate high current events
________________________________________
🔹 Supporting Observations for ECU Driver Fault
1.   Fuse blows only after engine is running → ECU transistors active
2.   Instant fuse blow if replaced after prior event → internal short is present
3.   Misfire codes appear on front bank → indicates driver not energizing coils properly
4.   Overheated old coils → consistent with overcurrent from driver failure
5.   Rear bank survives initially → localized to front bank driver
________________________________________
🔹 Likely Sequence of Events
1.   Original fault begins → ECU front bank low-side driver fails partially
2.   Coils on front bank slowly degrade from heat over years → overheating, swelling, cracked tops
3.   Once driver fails completely → fuse #24 blows while idling (dynamic)
4.   Engine loses front bank → runs on rear bank → rough, limp-mode-like
5.   ECU compensates → rear bank overloaded → fuse #2 eventually blows
6.   Engine light flashes, P0351/P0357 codes → OBD detects coil circuit malfunction
________________________________________
🔹 Summary / Diagnosis
✅ Primary culprit: ECU front bank ignition driver (low-side transistors) failure
✅ Coils: damaged by overcurrent in past, now replaced → will still fail if ECU not repaired
✅ Wiring / harness: intact, not the primary issue
✅ Grounds: secondary, likely fine
✅ Fuse #24: protecting front bank → blows exactly as designed, but indicates ECU fault
✅ Fuse #2: secondary effect of ECU compensation after front bank fails
________________________________________
🔹 Recommended Next Steps
1.   Do not repeatedly replace fuses → will blow instantly, risk damaging new coils
2.   Verify ECU front bank driver (optional, can test with multimeter for short to ground with fuse removed)
3.   Replace / repair ECU → front bank driver module is often not serviceable separately; requires ECU swap or professional repair
4.   Do not run engine on only rear bank → will overheat coils, stress ECU further
5.   After ECU repair → install new coils if necessary, fuse #24 should run normally
________________________________________
🔹 Optional Tests You Could Do Before ECU Replacement
•   Disconnect front bank coils → insert fuse #24 → see if it blows
o   If fuse holds → problem is likely coils or harness
o   If fuse still blows → confirms ECU driver internal short
•   Measure resistance from front bank coil positive to ground (with fuse removed)
o   Near zero → ECU driver is shorted
These tests are safe and confirm diagnosis without destroying fuses or new coils.
________________________________________
🔹 Final Thoughts
Everything you described — front bank only, dynamic at idle, instant blow on fuse replacement, overheat history of coils, P0351/P0357 codes — fits perfectly with a failing ECU front bank driver.
•   This is a classic symptom on older GD1 Honda i-DSI engines: years of ethanol running and normal wear could have stressed ECU drivers, eventually killing the front bank driver.
•   Old coils were probably victims, not causes.
So your root cause is not coils, wiring, or grounds — it’s the ECU driver internal failure.

________________________________________

[CRC]

In fairness, ChatGPT does make some interesting points.

I actually had one of those old Hondas with the faulty main injection relays ... a 1990 Accord that would start fine all year until the summer came and if I left it for a few hours in direct sunlight it would crank, but just wouldn't fire up at all. I found that if I could push the car into the shade and leave it for half an hour, it would then start first time.

Eventually a Honda technician I knew told me that the problem was poor soldering on the base of fuel injection relay and where to find it. I took it out and resoldered the pins at the base of the relay (very easy) and it cured the problem for ever. Took months to wok that one out .... why would the temperature of the cabin cause a problem??

My experience of alternators is that they either work or they don't.  Modern cars control the exciting voltage to the brushes to reduce sparking and extend their life, then the alternating current produced by the alternator (hence the name) goes through a ( 4 big) diode rectifier that turns the sine wave into a a pattern that looks a bit like the Lough Ness Monster .... in other words, DC voltage. This is then fed to the battery which is the ultimate smoother as it both charges the battery and supplies a lot of the current being used.

The Jazz has one more trick up it's sleeve with an extra circuit that will completely cut the exciting voltage to the brushes, thus not using the alternator to produce any power when it doesn't need it.
I didn't really believe this, but bought a small, cheap USB charger from eBay  that fits in the cigarette lighter which also has a voltage display in it, so that I can see the voltage when I'm driving along. Normal charging voltage is between 13.9 and 14.7 volts, but you will sometimes see it drop to around 12.5 when the power saving kicks in. If you then switch on the headlights, you'll see it start to charge again. Very clever and helps with the Jazz excellent fuel consumption figures.

I doubt if its the alternator as other things would be moaning ......... and possibly the charging light would stay red.

Is it the ECU?

Well, the coil switching is done by solid state switches in the ECU that's for sure, but if they were blown, then it would never run, and yours has run for 30 km or so without issue, so you have to question that assessment.

What I would do is find  the ECU (on a UK car it's down in the region of the passenger glove box I believe) and pull all the plugs off it, spray with contact cleaner and replace .... who knows.

An ECU from Honda would be prohibitively expensive, but do they have scrapyards in your neck of the woods? If your're lucky, you might be able to get one cheaply, but there are some differences between ECUs ... I remember a forum member who had bought a replacement Jazz engine, but when he installed it, it wouldn't run on his ECU, the reason being that the toothed wheel for the crank shaft position sensor had a different number of teeth!!

I think it's unlikely that the ECU is the problem .... but, who knows?

   

[Dwight]

Well I don't, that's for sure. Lol.

I don't know nothing anymore. Been scratching my head for 2 weeks now, car is out, can't use it and for the first time I feel like I've got some kind of issue I'm not qualified enough to solve myself.

And I don't have the money to call for a specialist.

So really, i don't know what to do, beside asking for an A.I scanning google for me for his opinion... Lol.

The only thing interesting about all that conversation, is the fact it made me realize both time it seemed to happen when car was parking and idling, but also heat was one. So while I doubt it has anything to do with the heat motor or assembly, maybe it could be something related to heat in the cabin in general.

I tested the continuity of the ignition relay, and that it clicks under 12V, but maybe that's not enough? Maybe it's bad like the one you're talking about once it's warm everywhere inside the car...

Could say the same about the ECU...

And the junkyards don't have Honda Jazz, or very rarely. I struggled already to find that engine when I changed it. Surprisingly, didn't change the ECU and it wasn't an issue. I guess they really found the exact same which is good.

So my only option would be to order it online, a used one, but they're not cheap at all, and nothing guarantees it's the issue. I'd like to be absolutely sure it's the cause of my issue if I ever ordered any. For now it's not even possible anyway.

Out of ideas for now

[CRC]

Measure those coil pack resistances and perhaps pull the plugs and just have a look. Should be NGK

I contacted an ex Honda technician but he'd never come across this issue in his time at a main dealers.

Unless you're very lucky to find a good one, most garages will employ the "scatter gun" method ..... replace everything, charge you for everything and their time, and then say "sorry, we couldn't find the fault"

You need to be like a terrier chasing a rat with these things ... it should start to feel "personal"

[CRC]

The above link is quite good as it describes how a 3 wire coil pack works.

Note train of 5 volt pulses coming from the ECU that causes a sold state ignitor in the coil pack to connect the primary coil to ground and thereby energise the primary coil.

At this point, the spark doesn't happen. It's only when the 5 volt signal from the ECU goes low that the ignitor disconnects from ground that the magnetic field collapsed, causing a high voltage in the secondary coil when then has many thousands of volts that will then jump the air gap at the spark plug.

Shows that a good ground  point on the system is essential.

[Dwight]

I'll try and mesure those coils (new and old ones) with a multimeter tomorrow if I can, we'll see.

About that ground, I didn't detect anything wrong with the harness ground wires, and I showed you the ground where all those black wire merge into one big one bolted to the engine, and it seemed fine, no rust, no green, tight...

However, I showed you the ground from the engine to the body itself (attachment 1) and while it does looks like verdigris made its way on the wire, it was not lose.

Do you know which relay is one for ignition coils? I know I pulled out thinking it was this one according to a quick google search, located above the fuse box inside the vehicle, but no guarantee it was the good one.

And if I'd find the one, how to properly test it? Beside continuity test and 12v supply "click" noise test? Should I open it and see if I see anything out of the ordinary?

[CRC]

I don't know where it lives on the Jazz, but I doubt if it's related to the problem. The issue with the relay mentioned by AI was that, when it got hot, it just wouldn't pull in when told to by the ignition switch, meaning no sparks and no fuel injectors, and obviously the engine wouldn't start.

I did check on eBay for an ECU price if needs be, and there are a lot on there, seemingly around the £30 to £40 mark, but no warranty obviously.

The reason I mention the ground points is that when the ECU switches the 5 volt signal to turn on the igniter in each coil pack, it's only route back is through the coil pack ground and then through the engine to chassis earth strap.

Clutching at straws a bit, but well worth a check ..

[Dwight]

Hello,

Had a bit time on my end today, could do some more testing.

So first thing, I rechecked continuity of all 12V and all grounds from each coil plug, up to the general engine bolted ground for the black wires, and up to the grey connector/plug behind the battery tray for the black with yellow stripe wires (12V). Everything was alright.

Then I pulled all the the "new" coils, and spark plugs for visual inspection and multimeter testing.

Spark Plugs:


Here's a look of two spark plugs (f'ront side, cyl 3 & 4 ): see attachment 1

Model: NGK V-Line 6465

Installed: Barely 3 years ago.

Gap inspection, all within spec (1-1,1 mm) but one which was under 1mm, probably at .9, reajusted.

Measurements with a multimeter set on 20k ohm

• SP1 (rear far left): 4.52
  SP2: 4.27
  SP3: 3.90
  SP4: 4.65
  SP5: 3.79
  SP6: 4.28
  SP7: 5.35
  SP8 (front far right): 5.18

Test on ignition coils (New)


First thing to mention, 2 of them are already showing sign of overheating, despite being barely used. Coil number 2, and more severe on Coil number 7 (again). Both showing swelling signs See attachement 2 and 3.

I measured on multimeter set on 200 OHM, inside coil plug pins. Pin 1 is far left (12V), Pin 2 middle (ECU), Pin 3 far right (ground). I did measurement between 1/2, 1/3, 2/3.

Results:
                1/2      1/3      2/3

Rear
Coil Cyl 1:  1.         1.        1.
Coil Cyl 2:  1.         1.        1.
Coil Cyl 3:  1.         1.        1.
Coil Cyl 4:  1.         1.        1.

Front
Coil Cyl 1:  1.         1.        1.
Coil Cyl 2:  1.         1.        1.
Coil Cyl 3:  82.5      83       0.7
Coil Cyl 4:  1.         1.        1.

Same test with old coils:


Note: ignition coils number 1/3/7/8 were showing signs of overheating and swelling.

Results:
                1/2      1/3      2/3

Rear
Coil Cyl 1:  1.         1.        1.
Coil Cyl 2:  1.         1.        1.
Coil Cyl 3:  1.         1.        00.7
Coil Cyl 4:  1.         1.        1.

Front
Coil Cyl 1:  1.         1.        1.
Coil Cyl 2:  1.         1.        1.
Coil Cyl 3:  A*        B*       00.7
Coil Cyl 4:  1.         1.        1.

A* when I put the second terminal on pin and closed the circuit, I immediately had about 100 ohm reading for a second then it droped to 1. Would jump again to 100 ohm each time I would remove the terminal off the pin and put it back.

B* exactly the same scenario as with 1* except readings would be between 50 and 170 upon touching the pin, then drop to 1. after a second.


That's about it for now, waiting for your feedback

[CRC]

That's brilliant information, well done.

I'm going to have a long hard look at that information and see if a pattern starts to emerge.

Couple of questions to clarify please.

You talk about cylinders 1-8 and also front and rear cylinders 1-4. Looking at the engine from the radiator end, could you clarify how the two descriptions relate to each other.

Where you show measurements of 1., presumably that means no circuit, ie. what you see on the multimeter when the probes are completely disconnected?

Interesting that the some of the new coils are also showing signs of overheating, and I've found out that can happen when the voltage is held on the primary coil for too long, causing it to "oversaturate" and therefore overheat and therefore pull large amps ....

Do you fancy a Zoom chat to talk this through sometime ... I have access to a zoom meeting place?

[Dwight]

Yes I should clarify this:

I number the spark plug and the coil packs the way, depending on their position,

So number one start on the very left of rear bank (first cylinder) and number eight ends on the very right of the front bank.

See attachment. I should have made it more obvious, my bad. From the start when I removed the old coil plugs, I took this habit of numbering them all this way, including the new coils and the old spark plugs.


What I show measurement of 1. it is indeed the same as if probes didn't touch at all, since when they touch they usually (at 200 ohm setting) show about .05

Although I would love to discuss this further with you like you mentioned zoom, I'm unable to discuss this orally because as you might remember, I'm not english native and it would be a terrible exercice for me. I'm more inclined to do the classic "typing" chat though, since I can at lest use a translator if needed this way.