Jump to content
Ultimate Subaru Message Board

Super Subaru Problem 93 Impreze with no spark


Recommended Posts

Hello all. What I've run into now is a 93 Impreza that won't fire. Here's the story. My wife was driving and the car began to over heat. She turned on the heater, and the temperature came down. She dorve on, then it overheated. she conitnued to run the car until it wouldn't run anymore, LOL. At which time I was summoned to fix the car. Well, after hearing the story, I figured I would find a siezed engine, but luckly the engine still spun freely and had no appearant water leeks, altough it was very low on water. I filled it with water then tryed to start it. The motor spun fine, but wouldn't start, so I checked the spark and found none. First, I checked to see if the coil pack was getting juce, and indeed the middle wire of the coil connector was hot, when the key was in the on and start position. Now here is where I get into the guess work, it kindof sounds like ECU failure, but how could that be related to the over heating? Also, I was hoping the car might be equiped with some sort of failsafe switch that would kill the spark if the engine was overheating. I searched all over for a reset switch and couldn't find a thag. Dose such a safty switch exist in Subarus? I figured that perhaps the ECU, might have such a system built in, and attemped to reset it by disconnecting the negitive batery cable, but that also failed to restore spark. I also clened all connection I could find under the hood incase water had gotten in. By the way, where would I look for the ECU, on this perticular car? I also checked all the fuses that I could find. To my knowledge there are 2 boxes 1 under the hood on the driver side finder, and the other under the dash. This is where I stand, a nice little car with no spark, and no idea why. Any ideas would be greatly apreciated, Thanks for your time, your new friend Jason.:banghead:

Link to comment
Share on other sites

With the low water and no apparent leak you probably blew a head gasket. Blown head gaskets have a few symptoms:

 

bad running

lots of sweet smelling water vapour in the exhaust

AND severe overheating (exhaust gas gets in to the cooling system, causing loss of coolant)

 

You probably now have coolant in your cylinders and contaminated spark plugs which will not fire.

 

DO NOT CRANK THE ENGINE unless you are positive that the cylinders are dry. Cranking a wet cylinder is bad, because water does not compress like air and you can break the piston, connecting rod, crankshaft or even blow the whole head off the engine.

 

You should pressure test the cooling system to determine if there is a leak.

 

You can also test the coolant for combustion gases (but this is usually done on a running engine to detect small head gasket problems).

 

You can also pull the spark plugs and see if there is water in the cylinders. If you crank the engine and water comes out of the spark plug hole, you have a blown head gasket.

 

Given that the car was driven overheated you may have done serious damage to the engine, such as warped heads or block, you wont know until it is pulled apart.

 

Subaru's have no "overheat" safety mechanism, other than to retard ignition timing and dump more fuel in to lower combustion temperatures. The only "overheat" systems are those on V8's that can run on alternate cylinders even if there is no coolant such as the Ford Triton and Cadillac Northstar engines (I still would not do this though).

 

Only when you are sure that the head gaskets are good should you look elsewhere for the problem.

 

You can determine if the ECU is working by noting if the check engine light comes on when you turn the key to "on". If no check engine light either the bulb is burnt out or the ECU has failed.

 

BUT I am 100% sure you have a blown head gasket based on the low coolant and overheating symptoms.

Link to comment
Share on other sites

I agree a head gasket sounds likley considering the circomstances, but I still can't figure out why the coils aren't fireing, could this just be chance? Or might it be related to the overheating? I also forgot to put in my last post, there is no visable sign of oil in the water, or water in the oil. Any ideas on the corolation between loss of fire from the coils, and the overheating?

Link to comment
Share on other sites

Depending on how you're checking for spark, you could have coolant (or vaporized exhaust valve metal) on your plugs, or a bad crank angle sensor, or melted plugs/plug wires or broken TB or ?

 

Start with what we know happens with severely overheated engines. follow alias's advice and confirm good compression first You quite likely could have holes in pistons, melted valves, scored cylinders, broken rings, warped heads, blown gaskets, and need an oil change. Spark is a secondary concern.

Link to comment
Share on other sites

When a head gasket goes, the high pressure within the cylinders when the engine is still running will more or less prevent a substantial amount of coolant from entering the cylinder. When you shut the engine off, the pressure in the cylinders falls to zero, while the coolant system still has about 14 or so PSI of pressure, thus causing coolant to get into the cylinder.

 

When you tried to restart the engine, coolant probably contaminated the spark plug. When coolant contaminates the spark plug, it reduces the resistance between the arc, causing a very weak spark, and also isolates the air fuel mixture from this weak spark.

 

Stick to the initial problem you found and find its cause first, in your case overheating and loss of coolant. Don't go off on a tangent checking coil packs, sensors, ECU, etc.

 

Not all blown head gaskets cause a visible amount of coolant in the oil, nor a visible amount of oil in the coolant.

 

Pressure test the coolant system, which involves pressurizing the coolant system to operating pressure (14.? PSI) and note how quickly the pressure falls (it should not fall if all is ok). This is often referred to as a coolant system leak down test. It will identify if the coolant system is leaking but not where the leak is. Since you don't have any sign of external leak, the only possibility would be a blown head gasket, or cracked head or block. There is an adapter which replaces the radiator pressure cap which allows the coolant system to be pressured. Once up to pressure let the system stay pressurized for 30 minutes and then recheck the pressure, it should not have changed.

 

If the coolant system pressure test is not revealing you can also perform a leak down test on each the cylinders to identify a blown head gasket. Sometimes this test is far more revealing because cylinder compression is between 160 and 190 PSI, much higher that the 14 or so in the coolant system. This involves the use of a cylinder compression gauge, and cranking the cylinder to TDC holding it there. Note the pressure and how quickly it falls off. All cylinders should lose pressure at a slow and equal rate.

 

 

BTW: A coil pack failure will cause a misfire code to be stored in the ECU, and it would in fact cause the engine to run rich and cool. A problem with the fuel system (pump, filter, injectors) could cause lean burn which could cause overheating (but a lot of engine knock first).

Link to comment
Share on other sites

yeah, but he says no spark at the COIL

 

there is a crank angle sensor near the front crank, it reads from the timing belt.

 

do you suppose you have a bad timing belt< and i its going out, was not turning the water pump properly,

 

i cant remember if the watr pump drives on the tooth side or the flat side of the belt.

 

suppost the water pump seized up, caising the motor to get warm, and then the belt break from froction on the pulley?

 

did you check for compression?

Link to comment
Share on other sites

Originally posted by MilesFox

yeah, but he says no spark at the COIL

 

there is a crank angle sensor near the front crank, it reads from the timing belt.

 

do you suppose you have a bad timing belt< and i its going out, was not turning the water pump properly,

 

i cant remember if the watr pump drives on the tooth side or the flat side of the belt.

 

suppost the water pump seized up, caising the motor to get warm, and then the belt break from froction on the pulley?

 

did you check for compression?

 

The water pump is "flat side" driven. I have never seen a EJ22 water pump seize. In fact if it seizes you will probably have the engine explode on the spot since it will melt and break the timing belt very quickly (perhaps less than 10 seconds), but that did not happen. The car overheated but continued to be drivable for some time. How much time?? Rubber belts do not like to move over stationary metal at high speed for very long.

 

The ECU will not allow the engine to run if the cam and crank sensors detect that the timing belt has slipped more than a couple of teeth. But the engine should have shut down immediately. In fact if the valves and pistons are out of sequence, the engine will barely run. If the cam belt sensor is defective or the belt is broken the spark plugs will still fire off the crankshaft reading. The ECU can fire spark plugs with either the cam or crank sensor not working, only with readings that indicate a belt slip will the ECU prevent firing.

 

He said no spark, not no spark at the coil. Perhaps a confirmation needed.

 

It is easy to check the timing belt, just pop the outer (left and right) timing covers off and rotate the crankshaft to TDC (align notch on crank pulley with "0" on timing mark on center timing belt cover. The camshafts should be at either 12 o clock aligned with the mark, or 6 o clock where you would have to rotate the crank one more time (2 crank turns to each cam turn) to align with the mark.

 

If the coolant system pressure test does not reveal anything, then you could check the timing belt before doing the cylinder compression test. Or check the belt first, it doesn't really matter as at some point it will have to be disassembled. If the water pump is bad coolant might leak out of it during the coolant pressure test, and if so the water pump will have to be replaced, and the pressure test performed again (you could still have a blown head gasket due to the overheating, or the water pump could have failed because of coolant loss relating to a blown head gasket).

Link to comment
Share on other sites

Hello all, here is the latest news on the mysterious overheating/loss of spark problem. After a long evening of detective work on the car, here is what I discovered. The water pump is indeed seazed. The timing belt is not borke, but the teeth around the crank are completly stripped off, allowing the crank to spin freely but not turning the cams. Here is what I suspect happened. The waterpump failed, but did not seaze at first, allowing the cams to keep turnig in time with the crank. As the car drove on, the failed water pump finally seazed, at which time the belt forze as well, and stopped the cams, thus, the cam position sincer quit sending to the ECU and the coils quit fireing. Oh, bye the way, I checked the spark by putting a screwdriver in the spark plug end of the spark plug wire and holding the metal shaft of the screwdriver close to the block, I did a compresion test and got all kinds of strange readings verying for 0 -110 psi, but since I later found the valves to not be moving that explains the failure of the compression check. Now the question remains. What other engine damage occured form the severe overheat? and how severe was the over heat really? Could it be the the belt borke just in time to keep the engine for self destructing? This all remains to be seen, I'll pressure up the cooling system after I change the water pump, and see if it will hold, I'm hopeful, there was no water in the cylenders, so that's a good sign, maybe I got lucky and dedn't blow a head gasket, or crack a head. I've also herd that some subaru are "Clearance Engines" meaning the valves won't reack the top of the piston, ever if fully extended when the piston comes up, so maybe I have hope there too. I'll keep you informed with the details of the adventure, and any advice, tips, or clarifications would be greatly appreciated, Later all, Jason

Link to comment
Share on other sites

sohc 2.2L MAY (do a search, there's still some controversy) be about the only engine that isn't interferential. Usually the engines that do exhibit interference are DOHC and it is valves that hit. But Emily says valves and pistons can collide on some soob engines - lots of conflicting opinions. I wish we had a list from FHI that we could 'sticky' because this comes up often.

Link to comment
Share on other sites

here is what you should do:

replace the belt and water pump. once the belt is installed, do a compression check.

 

if the compression is good, nothing to worry.

 

but getting the motor hot MAY cause the headgasket to be more prone to fail if it ever gets hot again.

 

if the motor never got too hot, it will be ok. the belt breaking as soon as it did most likely saved your motor.

 

if it were an ea82, you would have for sure cooked your heads

 

on my ea82, i got it hot once, off the scale in fact. but there was no damage. but it had fresh head gaskets before it got hot.

 

i ha e found that a new head gasket job will hold up rather well in a mild overheat, but an older gasket will want to let go before a new one would.

 

in my opinion, fix the belt and keep going. just consider a head gasket job later down the road, but it most likely wont be necessary now

Link to comment
Share on other sites

All Subaru EJ series SOHC are non interference and DOHC are interference.

 

Just curious, did your wife hear the squeal of a the timing belt on the seized water pump pulley? or was there no squeal?

 

I would have thought that a seized water pump would cause the timing belt to snap way before overheating takes place. It probably takes at least two minutes for the engine to overheat with a seized water pump, and I would have thought the belt would last maybe 10-15 seconds. I have have seized two water pumps on my 85 GL, and when they went the V belt broke before I could stop the car when I heard the belt start squealing on the seized water pump pulley. The belt went way before there was any increase in temperature. Of course it was a weaker V belt....

 

I have never seen an EJ series water pump seize, was this a rebuilt replacement water pump? I never use those......

 

I would install a new pump, timing belt and crank timing belt pulley. Be careful to inspect the cam belt tensioner, it may have been damaged. Change the crank seal and inspect the crank key and the crank accessory pulley to insure a tight fit, it is possible that they were damaged as well. The camshaft seals should also be replaced unless recently done.

 

It is actually possible that a defective cam belt tensioner caused this. I did see a case where a failed tensioner destroyed everything in the timing cover including the water pump.

 

Fill the engine with water and pressure test the coolant system for leaks, if all is ok, then run the engine. If there are bubbles in the overflow tank, you have a blown head gasket (the bubbles are exhaust). Run the engine for a few minutes after the thermostat opens (rad will suddenly get hot) to be sure. Stop the engine immediately upon the sight of bubbles. When the thermostat opens there might be a rush of bubbles, and this is normal but should stop in about 30 seconds or less (its just trapped air).

 

Note that you will have to open the bleed valve (left top of rad) to get the air out of the coolant system when refilling (on cars with bleed valves).

 

If you don't encounter any exhaust bubbles, put the car back together and run it carefully for a few weeks (don't take any trips where you could get stranded). Check the coolant level daily (or two or more times a day) to see if it changes, a loss means a head gasket is gone (unless there is an external leak somewhere). And watch for overheating and sweet smelling steam in the exhaust, both of which are signs of a blown head gasket. If you get through a month without blowing the head gasket, all should be ok (but you never know).

 

As MilesFox said, a compression test can reveal a blown head gasket, but it sometimes does not. A cylinder compression leak down test would though. Filling the cooling system and watching for bubbles is a sure fire method though, and you are not going to do any more damage than what has already been done.

 

The EJ22 engines have very reliable head gaskets, so it is possible that they survived. But you did seriously overheat the engine....

 

I have seriously overheated an older EA82 engine with no apparent consequences (it ran for at least another 200,000km). When my water pump seized, I drove it over 100km in 5-10 km bursts giving it time to cool down. It was -30 C out though, which aided engine cooling.

Link to comment
Share on other sites

Thanks for the info on the tencioner, I bet you may be on to something there, one ideler pully spilled it's guts and sent ball bearings through out the plastic timing belt housing. I'll definatly check the tencioner, by the way, what can you tell me about it?, dose anyone know what makes it work, it looked to me like a tiny hydrolic ram, dose if run off the oil pressure of the engine, or is it simply spring loaded? The cam seals are difinatly leeking like crazy, the inside of the hosing was soaked in oil. I don't know the history of the water pump, only had the car about a year. It very well may be a replacement, since the car has almost 250,000miles. Still was running like new though. I'll inspect the pump thourghly when I remove it to see if I can determin its orgin, and I'll let you know what I find. The timeing belt was slick with oil, and crunchy, I could break the teeth off with my fingers, can't believe it held up as long as it did. I don't know about a squesling sound form the best, she didn't mention it. What she did say though, that it was running quieter than normal, as it began to overheat, but can't say for sure about that, since I wasn't there. Take it easy all, Jason

Link to comment
Share on other sites

Originally posted by JasonGilman01

Thanks for the info on the tencioner, I bet you may be on to something there, one ideler pully spilled it's guts and sent ball bearings through out the plastic timing belt housing. I'll definatly check the tencioner, by the way, what can you tell me about it?, dose anyone know what makes it work, it looked to me like a tiny hydrolic ram, dose if run off the oil pressure of the engine, or is it simply spring loaded? The cam seals are difinatly leeking like crazy, the inside of the hosing was soaked in oil. I don't know the history of the water pump, only had the car about a year. It very well may be a replacement, since the car has almost 250,000miles. Still was running like new though. I'll inspect the pump thourghly when I remove it to see if I can determin its orgin, and I'll let you know what I find. The timeing belt was slick with oil, and crunchy, I could break the teeth off with my fingers, can't believe it held up as long as it did. I don't know about a squesling sound form the best, she didn't mention it. What she did say though, that it was running quieter than normal, as it began to overheat, but can't say for sure about that, since I wasn't there. Take it easy all, Jason

 

I don't now a lot about the tensioner internals, just that they are not 100% reliable. They definitely are hydraulic, but whether there is a spring in there as well, I don't know. The tensioner is self contained. BTW you will need to remove it, compress it in a big vise and insert an nail, drill bit, etc to lock it compressed (note the holes in case and plunger) when reinstalling the new timing belt. When you get the belt on, you pull the pin to tension, crank the engine twice by hand and confirm that the belt is properly aligned. If it takes a reasonable amount of effert (maybe 150-200lbs?) to compress it is ok, it should not be compressable by hand. All I know is that when the tensioners fail all hell breaks loose, and the tensioner will be very easy to compress, if it has any spring action at all, and often it is torn from the front of the engine (could be due to loose bolts).

 

When reinstalling the tensioner use threadlock on the bolts (and for all idler pulleys as well), I have seen these things come off even when torqued properly.

 

With the high mileage I would take the oil pump off and replacing/rebuilding it. The pump is most likely loosing pressure because of a loose back plate. Very easy to do. The oil pressure in my 93 Legacy was good at 466,000km, and I never needed to remove it, but I had an aftermarket oil pressure gauge to monitor for any problems. And I have not had to assist with the replacement on any of my friends' Subarus (34 of them and counting).

 

Someone else will have to add information regarding rebuilding or replacing the oil pump. I think there is an inexpensive kit, or the whole pump is inexpensive, all I remember is that it is not hard and not expensive.

 

Do you have any hydraulic valve lash adjuster noise (clicking)? If so it is likely the oil pump that is causing it (more true for old EA series engines, but also common on EJ series engines as well).

 

If your engine was quieter than normal, then perhaps normal was a rough sounding water pump!!

 

I am a little concerned that the idler pulley self destructed, not to expensive to replace, but is there any damage to the soft aluminum threads of the engine block?

 

Last time I got a new water pump it was $80 Canadian from a dealer, I was shocked how little it was. The reason I would not consider a rebuilt one it that it is timing belt driven and if something goes wrong..... While you've been down that path....

 

Rebuilt water pumps were at Canadian Tire (a combination Walmart/auto parts store here in Canada) were about $70, so not worth saving the $10.

 

How old was the timing belt?

 

Perhaps the oil contamination caused this failure, or a problem with a vibrating belt (due to bad tensioner) cause the cam oil seals to fail. They due fail on there own on a regular basis, but you've got "mega damage" to the timing drive system.

 

Be sure to clean out the timing belt cover area of all grease and oil with brake cleaner (because it leaves no residue) before you start to rebuild.

 

As for a replacement timing belt I recommend a Subaru belt since they have timing marks on them to line them up with crank and cam pulleys. I seem to remember that the crank pulley has two marks on it, its been a while since I did a belt, but I seem to recall an arrow on the crank angle sensor tooth was the correct one, not the arrow on the front of the pulley which is at 3 or 9 o'clock to the other mark. Someone please confirm this. Normally this would not be an issue since you don't turn crank or cam when just changing a timing belt, but you have to align from scratch.

 

As mentioned in another post, be sure to torque the crankshaft bolt properly to 130 ft/lbs, they have a habit of coming loose.

 

You can use a section of your broken timing belt to wrap around the crank pulley and lock it in position with a chain wrench to torque it.

 

Also be sure that the crankshaft key is in excellent condition (half disc of hardened metal which prevents the crank timing and accessory pulleys from free spinning), if in doubt, replace it. And insure that the slot for the crank key in the accessory pulley is not damaged. If the slot is larger than the key on the accessory pulley it will wobble causing the crank oil seal to fail, and damage to the oil pump and front crankshaft bearing. I've seen this happen a few times. Generally not an issue as long as the crank pulley was always torqued properly and the A/C compressor hasn't ever seized. I've seen 2 A/C compressors seize, and 1 of them caused the crank pulley to fly off denting the hood and puncturing the radiator....

 

Are you replacing the crank timing belt pulley, was it damaged (broken/bent crank angle sensor teeth)?. Also check the driver's side camshaft pulley's camshaft angle sensor teeth for damage.

Link to comment
Share on other sites

when squeezing the hydraulic tensioner, go slowly so you dont blow it out. put it in a vise and give it a crank till it binds a little, then let it bleed down. half turn, wait a second, half turn, wait a second, you get the idea.

 

i waspassing a semi in my ea82 powered 83 wagon, and it up and quit running. i pulled over at a hog farm to take advantage of the flood light and have a peek. i rin my timing belts exposed, no plastic covers. i could see that both ebelts were there. it confused me, so i cranked the starter from under the hood. i could see the crank turning, but not the belts.

 

it had stripped off likr 6 teeth, some were half peeled away, and the others i could pik off with my finger. abut 45 minutes later, i was on my way with a spare belt in the trunk(i had a spare motor i just picked up, too. not exactly the best scenario to be in going 100 mile home out of state on a suspended liense, in the same county i was arrested for driving while suspended on the way there the day before........

 

oh, the 93 impreza here in town is a 1.8 ej motor. 2wd 5spd

 

and as for thread locker on tensioner pulleys, u have had a pulley walk out enough to snap the bolt, because it was put togethe 1000 miles ago with anti seize. this happened 15 miles into a 2300 mile raod trip. then the axle broke in wisconsin. i still made it after my gas tank failed, but i got arrested on the way back

 

but i got my license now.....

Link to comment
Share on other sites

Originally posted by MilesFox

when squeezing the hydraulic tensioner, go slowly so you dont blow it out. put it in a vise and give it a crank till it binds a little, then let it bleed down. half turn, wait a second, half turn, wait a second, you get the idea.

 

i waspassing a semi in my ea82 powered 83 wagon, and it up and quit running. i pulled over at a hog farm to take advantage of the flood light and have a peek. i rin my timing belts exposed, no plastic covers. i could see that both ebelts were there. it confused me, so i cranked the starter from under the hood. i could see the crank turning, but not the belts.

 

it had stripped off likr 6 teeth, some were half peeled away, and the others i could pik off with my finger. abut 45 minutes later, i was on my way with a spare belt in the trunk(i had a spare motor i just picked up, too. not exactly the best scenario to be in going 100 mile home out of state on a suspended liense, in the same county i was arrested for driving while suspended on the way there the day before........

 

oh, the 93 impreza here in town is a 1.8 ej motor. 2wd 5spd

 

and as for thread locker on tensioner pulleys, u have had a pulley walk out enough to snap the bolt, because it was put togethe 1000 miles ago with anti seize. this happened 15 miles into a 2300 mile raod trip. then the axle broke in wisconsin. i still made it after my gas tank failed, but i got arrested on the way back

 

but i got my license now.....

 

Yep, go slowly on compressing the tensioner. Usually a vise is slow so I did not make this point.

 

And anti seize compound is the right stuff to use on unimportant bolts such as the timing cover bolts. All important bolts (such as the crank, cam, tensioner, idler pulley, etc) should be threadlocked. I use threadlock and anti-seize religiously and as a result I NEVER have a problem with seized bolts. When my 2001 Outback was new I took many bolts off and coated them threadlock and anti-seize so I wont have any problems 10 years from now.

 

It is also a good idea to check the crank shaft accessory belt after a 100 miles or so to see if it is still tight, just use a short wrench with a little effort (maybe 30 ft lbs) to see if it will rotate counter clockwise. I have had one that was properly torqued come loose after 100 km or so, but I had reinstalled a defective crankshaft accessory pulley (the slot for the crank key had worn to an inch wide!).

Link to comment
Share on other sites

  • 4 months later...

you would be surprised how long a timing belt will last when a water pump locks up! it will slide over that pump untill it melts the belt .

 

make sure the crank sprocket is on the 12 o clock mark [ which puts the pistons at the halfway point of the stroke] , turn the cams to the closed position [ cam turns freely by hand ], then use a cylinder leak tester to pressurize the cylinders to check the valves .

I have had a couple of 2.2 L engines break timing belts and not hit any valves.

make sure you use a subaru thermostat when you put it back together . have had way too many aftermarket stats fail .

the older 1.8 and 2.2 engines didnt have much head gasket trouble [ unlike the 2.5 ] .

the 93 didnt have obd-2 which wouldnt have the misfire detection .

the crank sensor handles the timing side and the cam sensor handles the injector firing sequence. either one failing will cause a no start.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...