[fifthwheel]

Hi. I'm new to diesel and to the i30 but heard many good things about it so I bought a 2008 1.6 CRDi with 87 megameters on the clock.

Problem:
White / blue smoke emitted from exhaust on start. Persists until engine is warm (about 30 seconds).
Also rough idle on start. Persists until engine is warm.

Observations:
80 amp glow plug fusible link blown.

Testing and diagnosis:
Checked continuity of plugs using multimeter set to measure resistance. All readings below 1 Ohm.
Checked potential between fuse box terminal (not connected to number 1 glow plug) and battery. Reading is 12.7 volts.
Glow plug relay tested for continuity using a 12 volt source to close the relay 'switch'. Relay performs normally.
Replaced 80A fuse (twice).
Connected relay pins to multimeter. See diagram.



Ignition switch set to 'lock', potential reads 12.7 volts. Switch to accessory (AKA acc) reading is 12.7 volts. Switch to 'on', reading is 0 volts. Stays at zero volts until switch to 'start' then reading is 13 to 14 volts. 80A fuse is hot!
Fuse may not last many more starts. See picture.



Thanks in advance for your help!

[Dazzler]

 

With all that info (well done) hopefully one of our honorary Technical advisors (or other friendly members) can advise.

I'll summon them and hope they are not busy!

@cruiserfied @nzenigma @tw2005

[tw2005]

can you energise the relay ans then measure the resistance across the contact side? Just wondering if the contacts are burnt or pitted and increasing in resistance drawing more current? Does the relay get hot too?

Nominal resistance for the glows in the manual says 0.25 ohms

[diesel1984]

Your problem is probably glow plug relay and burnt relay contacts as tw2005 said.

Resistance when measuring with ohmmeter can be in tolerance(0.03 ohm is ok when relay in triggered state) but in real life when glow plugs needs that current flow things are much different.

You can open relay housing and look at working contacts.

Burnt contacts looks like this:

[tw2005]

Your problem is probably glow plug relay and burnt relay contacts as tw2005 said.

Resistance when measuring with ohmmeter can be in tolerance(0.03 ohm is ok when relay in triggered state) but in real life when glow plugs needs that current flow things are much different.

You can open relay housing and look at working contacts.

Burnt contacts looks like this:

That's nasty. your photos? very  graphic

[cruiserfied]

Did you check that your glow plug zip rail is not touching the engine head? They can bend when tigtening nuts on glowplugs.

[The Gonz]

Not that I can visualise the zip rail right now, but the symptoms certainly describe a short circuit to ground after the relay.

[tw2005]

Not that I can visualise the zip rail right now, but the symptoms certainly describe a short circuit to ground after the relay.

[cruiserfied]

Not that I can visualise the zip rail right now, but the symptoms certainly describe a short circuit to ground after the relay.

I have absolutely no idea what the real name is. It's one of those things that i just say the first thing that comes to mind and hope that you understand because thats what we're talking about.
Zip rail
Buzz bar
Power rail
It all works lol.

[tw2005]

Not that I can visualise the zip rail right now, but the symptoms certainly describe a short circuit to ground after the relay.

I have absolutely no idea what the real name is. It's one of those things that i just say the first thing that comes to mind and hope that you understand because thats what we're talking about.
Zip rail
Buzz bar
Power rail
It all works lol.

Would you believe Glow Plug Plate? Someone was probably paid a lot of money for that name.

[fifthwheel]

Thanks for these suggestions. The glow plug rail did show some signs of 'pitting' in one place when inspected but I have bent the tabs to maximise the clearance from the engine block....  Glow plug fuse still blowing despite this modification.   
Next I will test resistance of glow plug relay - thanks for this suggestion!

UPDATE - checked glow plug relay contacts and found a minor amount of pitting - see pic (apologies for the clarity). 


Could a tiny amount of pitting like this cause the fuse to blow?
Now I'm thinking that there may be another underlying electrical problem causing the relay to burn out in the first place.
Question - Glow plug Relay is rated at 70amp - Glow plug Fuse is 80amp. Why is this so?

Thanks everyone for your help so far!

[nzenigma]

Ironically, I have just returned from a party ( south of Gatton) where it was 48 degrees C and due to be hotter today  .
And here I am trying to get excited about cold glow plugs. 

My best guess is that one of the glow plugs has zero resistance; it has a short circuit when 12v is applied.

Pitted relay contacts are inefficient. Therefore, they will reduce current flow not increase it. High current flow through the relay will cause them to discolour. Fifthwheel's photo looks normal.
Diesel1984's example is of damaged points that have ceased to be a conductor , the carbonised metal begins to heat and burn, increasing the damage and heat.

fifthwheel, remove the 'zip buzz bar rail' 
When you "switch to on" you should have +12v at the buzzbar connector.
If not , you have a short circuit between the hot side of the relay and the buzz connector.
If the buzz connection is +12v connect a heavy wire from that connector to each glow plug in turn. ( try a car jumper lead)
You probably don't have an amp meter so you will have to see if the fuse is getting hot with one particular plug. Also measure the voltage at each glowplug. Hopefully, the faulty one will be 0v. The others a low +ve reading.

Also check that the terminals of the fuse holder are clean. Carbon can cause a bad contact and heat.

 Switch to accessory (AKA acc) reading is 12.7 volts: You are reading 12.7v + on one terminal and virtually -ve on the other, this terminal goes to the glow plugs.

Switch to 'on', reading is 0 volts. : The points close, complete the circuit, 12v+ is lost across the fuse circuit.

 Stays at zero volts until switch to 'start' then reading is 13 to 14 volts. : Points open , engine is running so you are reading a higher charging voltage.

[fifthwheel]

Well, I removed buzz bar (plate) and tested each plug individually with some heavy duty (15A) wire that I'd bought to do exactly the test suggested by nzenigma earlier. The difference this time, was I trying to determine if the fuse was hot or blown whereas before, I was simply looking for a blown fuse.
The fuse did warm a little but certainly no melting of plastic and no blown fuse.
I replaced buzz bar and tried again and no hot fuse. No blown fuse either. I inspected the installation of the bar and the connection from the fuse terminal and it is exactly as I had it before. How do I know this? Because I inspected every millimetre with a concave mirror and a torch.
The damn thing has fixed itself! Car starts with no smoke.
The only thing different was I took apart the relay, smoothed the contact surfaces with a piece of paper (closed contacts with fingers and gently dragged paper through) and wiped contacts with microfibre cloth. I also tried to gently bend the contact arm so that the two contacts might meet in a slightly different place. I have no idea if the bending did anything as the difference before and after was imperceptible. I'm doubtful I did anything of consequence.
Anyway, I got the idea from this page, Relays failures.

I strongly suspect my victory will be short lived.

Thanks again for all your help.

[tw2005]

Well, I removed buzz bar (plate) and tested each plug individually with some heavy duty (15A) wire that I'd bought to do exactly the test suggested by nzenigma earlier. The difference this time, was I trying to determine if the fuse was hot or blown whereas before, I was simply looking for a blown fuse.
The fuse did warm a little but certainly no melting of plastic and no blown fuse.
I replaced buzz bar and tried again and no hot fuse. No blown fuse either. I inspected the installation of the bar and the connection from the fuse terminal and it is exactly as I had it before. How do I know this? Because I inspected every millimetre with a concave mirror and a torch.
The damn thing has fixed itself! Car starts with no smoke.
The only thing different was I took apart the relay, smoothed the contact surfaces with a piece of paper (closed contacts with fingers and gently dragged paper through) and wiped contacts with microfibre cloth. I also tried to gently bend the contact arm so that the two contacts might meet in a slightly different place. I have no idea if the bending did anything as the difference before and after was imperceptible. I'm doubtful I did anything of consequence.
Anyway, I got the idea from this page, Relays failures.

I strongly suspect my victory will be short lived.

Thanks again for all your help.

I think get a new relay because it sounds that's the change. Would have liked a resistance reading before and after on those contacts. I may have a quick peek at one I have here and check out the contact surface.

[tw2005]

Looking at one with 41000k, it has something similar to yours, I'll try and photograph it.

Yeah, I don't know what to think TBH.

[nzenigma]

Hey great news.

You sound doubtful. Lets work on that.

1. we know the glow plugs are ok. Great
2. Re my post, poor relay contacts cause low or no current to flow, NOT high current.
3.  We know you had an issue with high current because you blew a fuse.
4. Your new fuse(s) were still getting hot; therefore, still high current.
5. You mention some pitting on the buzzbar. If it was touching the engine block, it will be pitted by the short circuit and it will blow a fuse.
6. The only way I can think of your relay blowing the fuse is if the internal +12v contacts were loose and they touched the -ve side of the electromagnetic coil. However, an 80 amp short will leave a lot of soot and molten copper.

[The Gonz]

If it stays good, it's likely some short caused by some oxidation or other debris between the rail zip bus bar plate and its mounts, which was dislodged be removal and refitting. We may never know.

[fifthwheel]

Yeah - seems as though it's something electrical alright - a short somewhere. 
Pretty sure the first fuse blew because of this shorting evidence on the bus bar (see pic)



We'll probably put a new relay in just to be on the safe side and see how it goes from there - keep you posted!

[nzenigma]

Hmmmmm Looks like 80 amps to me 

You can imagine what the inside of the relay would look like. Splattered!

It wouldn't surprise me if you find something metal, like a bolt, is missing from somewhere above the rail.

Would be worth dropping your plastic sump protector to see if anything is in it.

[fifthwheel]

Is it possible - because I've tried hard to investigate other possibilities - that a relay could 'weld' itself into a closed circuit? Temporarily?

I am convinced that a fuse can only be blown if the circuit essentially has no resistance. And it's easy to blame an uninsulated positive buzz bar. So I've scruntinised this thing to the very last atom. (BTW, Why make it this way?)

I've had some very strange results including the engine stalling (twice!) after replacing the first fuse. I realise I'm new to this, but there is definitely some voodoo magic thing going on to now have it working just fine. I should add that I donated a 1997 Toyota to a good cause so the Car-Gods smile upon me, and this might explain something.

Thanks for all your posts. If the i30 mojo runs out again, I'll add to this thread.

[The Gonz]

Yes, relay contacts can become sticky when they deteriorate, although I'd expect heavy duty ones to have a good strong spring return action.

[tw2005]

Is it possible - because I've tried hard to investigate other possibilities - that a relay could 'weld' itself into a closed circuit? Temporarily?

I am convinced that a fuse can only be blown if the circuit essentially has no resistance. And it's easy to blame an uninsulated positive buzz bar. So I've scruntinised this thing to the very last atom. (BTW, Why make it this way?)

I've had some very strange results including the engine stalling (twice!) after replacing the first fuse. I realise I'm new to this, but there is definitely some voodoo magic thing going on to now have it working just fine. I should add that I donated a 1997 Toyota to a good cause so the Car-Gods smile upon me, and this might explain something.

Thanks for all your posts. If the i30 mojo runs out again, I'll add to this thread.

Do you mean randomly stalling , like driving along and it just dies or difficulty starting?

[Phil №❶]

Not likely to be welded contacts on relay as he mentioned that when working, "car starts with no smoke"

[nzenigma]

Is it possible ...- that a relay could 'weld' itself into a closed circuit? Temporarily?

 Unusual but, Yes. However, the contacts would have signs of 'welding'. Yours did not.

I am convinced that a fuse can only be blown if the circuit essentially has no resistance.

Most of an electrical circuit will, and must have, little to no resistance; eg wires, fuses, relays, switches, etc.
We use fuses because we know things go wrong. The fuse is an essential safety device, incorporated into any electrical supply line. Its function is to blow open if there is excessive and damaging current flowing through a circuit. (eg. a buzz bar short circuit [= 0 ohms])
 The fuse is a simple component that has been saving lives since Louis Breguet grew a beard.

And it's easy to blame an uninsulated positive buzz bar. So I've scruntinised this thing to the very last atom. (BTW, Why make it this way?)

Very easy to blame it,  you seem unconvinced ????, even though there is physical evidence that it has been short circuited ( possibly by some foreign metal object) to the engine block.

The glow plug fuse protects the glow plug circuit only.
Logic dictates that an 80amp fuse could not protect more delicate electronic components. Therefore, the stalling (so-called) has no relevance to this RARE malfunction of the buzz bar rail strap thingo.

[fifthwheel]

Alas, my problems have not gone away. Engine starts just fine but I checked the fuse again and was mostly impressed by this mutant Shrinky Dink (see photo).



Is there some kind of controller? A PCM that should open the circuit to prevent this? A 555 timer 555 timer IC - Wikipedia in the ECM perhaps?

Manual says to wait until dashboard glow plug light is out. But maybe I should wait not a second longer before I turn the engine over? I'm at risk of starting an engine bay fire if the fuse doesn't do what it's supposed to do.

Please help me to not blow my car up.

Thanks.

[Phil №❶]

You need to find the short 

[nzenigma]

The glow plugs will remain 'live' for a period after the engine starts. Cant say how long, doesn't really matter, because the fuse is capable of handling it. If you think the glows may be on permanently, meter the buzz bar when motor is hot.

[Phil №❶]

If you disconnect and isolate the feed wires from earthing, does the fuse blow at any stage when ign is switched on. Even when cranking the engine

[nzenigma]

Hi Phil,
Yes connecting wires may have been cooked and there is an intermittent short to ground due to bad insulation.
The story has unravelled that the buzz bar was not shorting. But, then it was revealed that a chunk is missing from it indicating that it had been shorted. The fault then disappeared. Irrespective of all the above flaky science, it would simply appear the bar shorted and blew a fuse but may have damaged some insulator. Vibration can make the short come and go.

[fifthwheel]

It is true that the buzz bar was touching the engine block and I was certain that this was the reason why the first fuse blew.
Why the second blew; why the engine stalled and why the third fuse melted the casing is still a mystery to me. I do like the hypotheses that grounding has occurred - damned if I can find it.

Below I've posted a video (apologies about the quality, I had to shrink it to fit Aussie bandwidth) that shows a new fuse and the relay in place without its cover and with me turning the ignition to the 'on' position and leaving it there. You can see at the 3 second mark that I've turned the ignition to 'on': the relay contacts close. At about the 18 second mark the contacts open.

What is the timer or magic box that does this?

...The ignition is still in the 'on' position beyond the 18 second mark.

As you will also see from this video, nothing is wrong! No heating of the fuse and the relay appears to be working fine (except I've removed the cover and I don't know what the trigger is to open the circuit).

So it appears to have fixed itself again! I need to give up and visit an auto-elec don't I?