All coils fired at once. Engine blew?

[mxp_57]

Hey guys,

I am building a Nissan s14a swapped with a 2jz-ge vvti + turbo. Using ms3x and ls2 truck coils. It was running last summer but i was finishing the build and upgrading headstuds/freshing up engine.

When trying to first restart the car in cold weather using the old map it wouldn't start as the water temps was reading wrong before and therefore maps were now off. As i was trying to get it to start (injectors in batch mode for now to get it to start) the engine got flooded. I pulled the fuse to the fuel pump as we wanted to check firing order (should have used the test mode in tuner studio...).

But the weird problem happened when the megasquirt was ON and the fuse to the fuel pump only was removed (coils and injectors are still powered), as i turned off the MS power to restart after changing settings it fired all the coils at once...! Causing a really really loud detonation with the remaining flooded fuel in there.

After that the compression seems leaking in one cylinder to the breather system so i most likely broke something from a few cylinder igniting strong at the wrong time. I'm surprised i broke something with the engine not running but i can imagine it being possible. And i can hear a soft TONG TONG TONG noise like coming from one cylinder.

I'm going to test compression and inspect everything but my main concern is what could have caused this issue so it won't happen again? Should the power to the coil be turned off first before turning off the megasquirt or is it a wiring issue or ground issue causing this firing when turning off megasquirt?

Also when the fuse to the fuel pump is in place i can turn off and on any switches it won't fire the coils. It's only when the fuse to the pump is removed.

I'm using a fuse box and wiring similar to the diagram to power the coils, injector and MS separately but with switches (no relays or ignition key). So instead of the fuel pump relay, there is one switch for the MS/electronics and another switch to power each independent fuse for fuel pump, injectors and coils (like on the MS3 wiring). And fuel pump is grounded to the chassis. It's only when i removed the fuel pump fuse and and with the switch ON to the coils / injectors, and then turn off the megasquirt that it fires all the coils at once.

The coils have the power ground on the engine and the sensor ground going back to the megasquirt. Then all megasuirt grounds are going on the engine.
The battery is in the back, grounded to the chassis at the back (same as fuel pump) and grounded at the front on the engine and to the chassis.

Thank you everybody for your help.

[mxp_57]

if using the relays like the MS wiring diagram with the ignition key switch powering main relay, then fuel pump relay triggered by megasquirt then the power will always go off on both at the same time. So the coils cannot stay on while the ms goes off. Could that be the reason why? But it still doesn't explain why why it fires all the coil only with the fuel pump fuse off...

[billr]

Let's start there...

[DaveEFI]

Since you have independent switches, is it possible to have the coils powered up when MS isn't? Or does the main MS switch also power the relays (via their switches) for everything else?

[mxp_57]

I read about the msq file before posting but i didn't think it was relevant here. It's more of a wiring/MS3 behavior issue. (correct me if i'm wrong) I get no sync loss and car was driving great before with this tune.

Yes the MS can be powered alone without the coils being powered and the other way around too as i use a small battery and i wanted to power it independently when looking at logs etc.

Basically there is the big battery isolator switch, sending power to the switches, 1 switch for megasquirt and 1 switch for coils, inj, and fuel pump al together but with separated fuses. (like on the MS3 wiring drawing but with direct switches instead of relays triggered off ignition key and MS.

So there is one wire going from the coil/inj/fuel switch and powering a few fuses one for fuel pump, one for coils, and one for injectors.

But it doesn't fire the coils when turning off MS switch unless i remove the fuel pump fuse.

I am still unsure if the cross firing could have bent a rod or blow out the valve seals on one or few cylinders, maybe damage pistons or ring land but something is wrong. I will pull it appart this week to check it out.

I know how to assemble a MS or wire a car etc, but i'm not really expert in basic electricity science so maybe i did a stupid wiring mistake trying to make things simpler and the layout is not correct i don't know. Here is a pic of the engine, as you can see i like things clean and simple.

Thanks!

[jsmcortina]

I am still unsure if the cross firing could have bent a rod or blow out the valve seals on one or few cylinders, maybe damage pistons or ring land but something is wrong. I will pull it appart this week to check it out.

Just as a thought - compare the forces inside the engine when there's a little fuel in the cylinders and the valves are open to when the engine is running at full load.

At full load you are compressing the charge in the cylinder to say 9x atmospheric and then exploding the mixture. Here you were at atmospheric and the valves were open - far, far less force on anything.

Also, make sure that your coils can NEVER be powered when the Megasquirt is off. Ideally power them from the fuel pump relay as the diagram shows in the manual.

James

[mxp_57]

Thanks James for the help as always.

I will remove the coils from the switch and but them on a relay. And keep a switch for the injectors and fuel pump. I thought maybe a grounding issue would cause something like this. I don't understand how it can trigger the spark to turn off the MS. Does turn it off cause a similar current situation as when the MS triggers the coils?

I know it seems weird to blow an engine at cranking load but running e85 and with the large amount of fuel that was sitting on the pistons it might have exploded strong on one cylinder at tdc with valved closed and tried to push down the pistons while other cylinder were also lighting up fuel? The explosion was really really loud and internal. There's seems to be fuel in the oil and the oil now turned milky color too. Also it feels like compression is now leaking in the crankcase ventilation and it seems to be the weird sound i hear now from the engine when cranking after the backfire.

[kjones6039]

the oil now turned milky color too.

Pretty much a sure sign of coolant!

Ken

[ugp]

If a backfire has caused that then you've had some underlying engine issues from the outset. The pressures exerted in those combustion chambers is much higher either under boost (as James has already mentioned) or when you bash it off the rev limiter. Doing the latter is exactly the same as igniting the fuel sat in the bores. The fact that you heard it as a loud bang would have been due to the exhaust / intake ports being open at the time. Had the valves been shut on the cylinder it caught on, the crankshaft would have rotated a few degrees and you'd have barely noticed it.

It sounds to me like you had a head gasket on the way out. These engines aren't bomb proof and I've currently got two in for blown heads on stock engines. They're also getting on for over 20 years old. Hell, Rover didn't get 20 days out of a head gasket, never mind 20 years.

Ben

[mxp_57]

Thanks for the reply.

Well i'm quite stumped right now. I don't think the backfire caused anything. I have massive compression problem. To give you a bit more info, i bought this long block second hand and when i got the swap running a few month back I did some quick test and tuning and the engine ran great for driving/cruising and moderate acceleration and datalogs were great.

So i did a few boosted pulls from 2psi up to 6psi. It was running really strong. After the first 4 or 5 quick WOT pulls the engine started blowing oil from the breather and ran rough. It could still start when pushed and drove back to the garage. The next day i tested compression and it was something like:

1: 30psi
2: 35psi
3: 40psi
4: 60psi
5: 100psi
6:160psi

(ps: it's not the battery dying as it gives same results when starting the test from 6)

I had performed a wet test and it didn't raise. Rods were not bent too. Some of the valve seals were shot but i thought it might have also been a head gasket failure or also maybe the head bolts stretching. I got the head resurfaced and new head gasket 0,6mm instead of 0.3mm and ARP bolts (lowering the compression a tiny bit). I changed the valve seals, valve job and adjusted clearances with new 264 street cams.
The head was torqued properly. New timing belt and the timing is still correct.

Now yesterday i did another compression test and i got pretty much the exact same results as before doing all that...

I don't know if it's a cracked block or cracked/warped head. I regret not having the head pressure tested before. I'm going to get a leak down tester to help diagnose further. I had bought this engine second hand so i'm not sure what happened in his past life. When i removed the head to replace the HG the last time (according to the toyota procedure/order), some bolts were craking strong while some were loosening easier.

I hadn't turned on the WI yet but intake temps were dropping on accel and only 65C degrees on boost. The AFR was in the 10.5/11.5 range. I had no signs of detonation (i'm using a J&S safeguard also). Last time, i thought that the timing might have been a little too advanced. I used a map from a na-t but i learned later on that my NA vvti had 10.5:1 compression intead of 10:1 for the non-vvti NA. The engine was running cool, no overheating.

I know these engine aren't bomb proof but i can't say if it is bad luck or not as it doesn't seem like i did anything harmful to crack or warp the head/block. Could the vvti failing cause these sort of compression issues? I don't think so but i really can't figure out what happened. Any help is most welcomed!

I still have the log file somewhere i think from when it blew the first time but i spent time looking at it and couldn't figure anything wrong in the different temperatures or AFR etc.

Thank you!

[billr]

With the head off: put a cap, a fake "head" made of aluminum or thick (2" or more) plywood on one cylinder. Seal it to the cylinder with a rubber sheet for a "head gasket" and then try applying air pressure. Do that for each cylinder, one-at-a-time. Let's see if the cylinders/pistons hold air. You should be able to pressurize them to the 100-150 psi limit of most shop air-compressors that way.

[jsmcortina]

Even easier than that - just listen while the leak-down tester is on - you'll hear where the air is escaping.

James

[mxp_57]

Yes that's what i will do. I have ordered a leak down tester. I might as well keep the head on there until i can figure out where it's leaking from.

[WillExoIX]

Perhaps from cranking the coils were still energized, and when you cut the power it broke the magnetic field and fired the residual charge in the coils?

Thats all I got and i'd like to consider myself relatively intelligent.

Will the coils even hold the energy if switched 12v is on but no longer cranking/running? I would assume there would be some voltage/current bleed once no longer cranking.

But what jumps out at me, is that the magnetic field would collapse once power is cut leading to energy discharge (spark), and with no cam/crank input it fired all at once?

[mxp_57]

I was imagining something along those lines. But still the weird thing is that it's not just some field coming from the ecu only. It is when ecu is switched off and the fuel pump fuse which is wired on the same main 12v as the coils switch is removed... I'm guessing putting a relay instead of the ignition/fuel switch might cause the same triggering of the coils if the fuel pump is removed (or blown if considering possible failures) and then ecu is switched off.

I will do some more tests before i hook back up the coils, i might just have to always turn off the ignition switch before the ecu.

Edit: on a side note, it was a pretty big spark. I removed the spark leads and the coils are mounted 10cm away from the shock tower which is painted and we did some test with the fuel pump fuse off and switching off ecu and it was making arcs from the coils to the tower. (from the last coil mainly which was the closest to the tower.)

[billr]

"I'm guessing putting a relay instead of the ignition/fuel switch might cause the same triggering of the coils if the fuel pump is removed (or blown if considering possible failures) and then ecu is switched off. "

The difference would be that a relay controlled by MS detecting cranking wouldn't be energized until the MS had been fully "powered on", and the spark drivers then have been set to the proper state for keeping the coils "off" until a spark event is required.

[WillExoIX]

I was imagining something along those lines. But still the weird thing is that it's not just some field coming from the ecu only. It is when ecu is switched off and the fuel pump fuse which is wired on the same main 12v as the coils switch is removed... I'm guessing putting a relay instead of the ignition/fuel switch might cause the same triggering of the coils if the fuel pump is removed (or blown if considering possible failures) and then ecu is switched off.

I will do some more tests before i hook back up the coils, i might just have to always turn off the ignition switch before the ecu.

Edit: on a side note, it was a pretty big spark. I removed the spark leads and the coils are mounted 10cm away from the shock tower which is painted and we did some test with the fuel pump fuse off and switching off ecu and it was making arcs from the coils to the tower. (from the last coil mainly which was the closest to the tower.)

Well that would explain why it happened. The coils were energized and the field was being held until you pulled the fuse and cut the power. It only had one place to go.

One of the many reasons IMO fuel and spark should have separate fuses/power sources.

Mine have separate power sources. (Diff fusebox location.)

[ugp]

What usually happens with the 2jz (I've got one in at the moment with this, and had two in this year the same also) is the valve stem seals leak. People ignore them, to the point that the bores glaze up. Result of this is loss of compression and occasionally (as the one in at the moment does), is push oil up past the rings.

Only fix is a rebore and new rings.

[mxp_57]

Thanks. Is it possible that the head gasket failure across cylinders caused the compression to leak and blew the valve seals during the few pulls to test at wot? And driving back 10km back to the garage with heaps of blowby and burning oil could have caused the rings to stick or wear and/or cylinder walls glazing? it didn't overheat or loose coolant and the oil and coolant didn't mix.

I checked the leak down and it's leaking badly on 5 cylinders into the crankcase out from the oil cap and breather hose. And a little bit from the oil dip stick hole. cylinder 6 is holding pressure nicely.

The engine wasn't started before so the valve seals were probably a bit worn but the car wasn't smoking at all when i was tuning the idle and cruising/light load. It blew after 5 or 6 pulls up to 6psi with the wastegate not closed fully so boost wasn't coming on strong. I stopped after realizing it was blowing oil out everywhere. AFR was ok. I'm not sure if the timing map was correct.

I have the catalog from when it blew but it's 4mb, how can i compress it to upload it?

Edit: could the rings have gotten stuck in the pistons from burning oil and fooled plug?

[billr]

Whatever it is, the head probably (must?) has to come off to fix the problem. Do that, see what failed, and we can all stop wondering.