Diagnose an explosion?

Smoke from burnt wiring somewhere?  No heat so maybe just current through a wire?.

Ya think mice got in exhaust system in that month?  Not much wiring under there to causes smoke - just oil pressure sensor.  Check oil for gas smell and overfill condition. 

     Not burnt wiring Frank.

There was far, far, too much smoke for a briefly smoldering wire, and all that smoke was preceded by a very startling definite explosion

     No mice Greg. I liberally lace the VS with mothball bags.

The smoke wasn't exiting through the exhaust. It was curling up from under the tin into the engine bay through the opening in the protective semi-circular crank/timing pulley.

     I believe I know what happened though:

Pumping the pedal dumped raw fuel into the cylinders. that accumulation of excess fuel and fumes finally ignited causing the explosion. Why it didn't vent through the exhaust is a puzzle unless the exhaust valves were all closed...it must have vented from the crank case out through the pulley attachment (and that sounds serious!)

     I'll read up on how to determine acceptable end play for the crank. If that seems OK I'll examine the plugs (and wonder why they weren't blown right out of the heads!) Then check valve gapping and the timing. 

     If that's all OK and it still idles rough I'll have to pull the rocker arms off to see if any of the push rods are bent...if not, The heads will have to come off to see the surface condition of the pistons.

     At this point I've gone as far as I can go.

Without a scissors lift I'll probably be crippled anyway!...If there's an easier way to diagnose and fix this I'd love to hear it. 

A backfire that popped an intake gasket would do what you are saying. It would also account for the smoke in the cylinder tin that came out of the gasket leak.

The "backfire" scenario is very likely the problem....   The least invasive first diagnosis would be a compression and bleed down check.....  At least it might localize the problem cylinder....   

     Thanks Leon...Your suggestion of the least 'invasive first' (compression and leak down) is exactly the advice I'm after. I'm open to any and all 'experienced attained advice and insight' into this...especially when I so sorely lack such experience!

     I was up early this morning following the lead that possibly a backfire blew out an intake gasket. Difficult to inspect but using a flashlight and a small diameter articulated mirror I couldn't see (or feel) any such indication between the mating of the heads to the intake manifolds...but I'll look up the torque specs for the bolts and check them out even though I doubt that caused the problem because the smoke filling up the engine compartment was definitely rolling up from under the tin through the slot surrounding the crank/timing pulley... and not escaping from the base of the manifolds. 

Carl....   If it was a "backfire", the manifold gasket is probably history.....  Quick check is the old spray the manifold base with choke/carb cleaner with the engine idling....( a change in engine running indicating leaky gasket. ) Spray the carb base also.(same reason)

If the gasket is shot (either one), check the studs when the carb/manifold is removed....  I had one that pulled out the threads in an old single port head....

Good luck...

Leon is giving some great advise, but it will probably take a few defined steps to get you there while you use his advice, so, let's review the upcoming steps to solve your mystery, knowing that it seems to run at higher RPM's but stumbles around idle.  Try to do these more or less in order:

1.  Get it running and warmed up a bit so that it idles as best it can.

2. While idling, liberally spray the base of the intake manifolds with carburetor cleaner - use the aerosol can (Gumout, NAPA, Sea Foam, any of them will work - the cheaper the better) with the little spray tube attached.  Don't forget to spray the back side, closest to the head tins.  If you notice ANY change in idle lope while spraying a stream at the intake base you've got a leaking gasket.  Give it a good blast (1-3 seconds) on all sides and try to get the output end of the tube close to the mating surfaces.  If there is a leak, the idle note should change - noticeably - and it should be repeatable with additional sprays.  This test will usually NOT find a leak between intake ports, where the gasket is thinnest, but it's worth a shot (pun intended).  There is no easy way (at least that I know of with these dual carbs) to do a manifold vacuum test so that's out.  All you can do is throw up your hands and swap out the manifold gaskets.  

3.  If nothing changes in #2 above, do a compression test.  If all cylinders are above 110 lbs. and are all within 10 lbs of each other, you've got a good engine.  If not, you now have something more to look at and could do a leak-down test to pinpoint the problem cylinder(s).

4.  If everything in #'s 2&3 above is hunky-dory, and because of your other symptoms, I would pull the idle jets and clean all of the jets.   Just as well, remove the mixture screws (don't lose the springs or o-rings (if it has o-rings) and use that same aerosol carb cleaner and blast into the mixture screw hole for 2-3 seconds.  If you can muster 40 lbs. of compressed air, you could blast that up in there, too.  Put them all back together and see what that does.

It's not unheard of for a big cough up through the carb to sound like a big BANG! and have it slam some crud up into the idle jets to clog them up.  That would be my first place to look, but please go through the other 3 steps, too, for your own peace of mind.

If it was a backfire (on the exhaust end) and it caused a momentary pressure spike in the crankcase, bear in mind that there is NO oil seal of any kind on the pulley end of the crankshaft.  There is an "oil slinger" inside the case, which is simply a washer on the crankshaft and is designed for the oil to hit it as it works it's way along the shaft and the spinning of the shaft causes the oil to sling off of the washer and drain back into the sump.  Since there is no seal, a pressure spike can cause a belch of oil past the slinger, out the back of the engine and, potentially, onto the exhaust pipes back there under the pulley.   THAT would cause a cloud of oil smoke to rise up from under the tinwork.  Poking around under there with a flashlight should show some evidence of that.

If all THAT doesn't fix things, you may need a Piperato/Cuccurreddu/Nichols Intervention Team visit.  Who Knows?  Might even be able to get "Sand-Bar Al" to come along, especially if Pearl is running by then.  

This all sound good Leon??

the biggest downside to a Suby engine is that if something were to go wrong (... God forbid)  .... no way I could get this kind of group advice and support.  You guys are great

But most major towns have a Subara dealer or mechanic who could quickly figure out what is wrong with the engine.

Not many places have a shop that knows diddly about aircooled engines. 

Reads good to me Gordon.....  I'm never quite sure how capable some owners are....

Yup, not a lot of Aircooled support left out there, and what's there is often dodgey at best.  

LOTS of Subaru support, plus lots (most) of the Suby-Speedsters have that much-coveted OBDII port to help your diagnosis.

And besides.......who says some of us don't speak "Subaru"??

Carl,

Your car backfired hard. Jack it up and check for a muffler that's split. I dropped a distributor in 180* off once, and had a backfire so bad it shredded a muffler. No harm whatsoever to the engine, but it freaked me out pretty badly.

All of Gordon's ideas are good, but I'll bet you find that nothing is worse for the experience. If you blew out an intake gasket, it was likely a problem anyhow.

I'm betting it'll be fine. Minor surgery at the worst.

Gordon, I'm up for a Carl Berry rescue mission, keep in touch and we'll get it done.

Danny & Gordon...you guys have a great heart!!!

But first give me the opportunity to follow the step-by-step advice and fix it myself...I need that experience!

If I hit a stone wall I'll then fall on my knees beg for a 'rescue mission'

Tiger at A-1 Muffler gave me a way to check for leaking intake manifold gaskets and that is to obtain a length of clear plastic tubing---like you see used in aquarious but about 1/4" diameter.

Warm the engine to opeating temp then when it's ideling put one end of the tubing up to your ear---sort of stick it in there and the other end to different areas of the exhaust pipe where it goes into the engine.  If there is a leak you will hear it clearly loudly and know the gasket is shot.

I haven't tried it but I filed it away in my "knowledge files" for when the day comes.

     OK people, It seems dramatic but I've come to the conclusion that there was a fire in the crank case because all the smoke was billowing up through the narrow slit in the tin surrounding the power pulley. None was it was coming from the upper engine compartment...and the smoke was being produced for several seconds. Far more smoke than just the result of one sudden explosion.

     The fire must have been burning off raw gas dumped into the crankcase (stuck valve?) until the available oxygen for combustion was depleted and it smothered itself out.

     But HOW was the smoke escaping from the crankcase?

     The only explanation that makes sense to me is that the pressure of the explosion affected the integrity of the bolted connection between the pulley and the cam (allowing the smoke to escape) andaccounting for the subsequent ragged idle, vibration, and ratty noise.

     I'm thinking that the first thing I should do (now that the rib cage seems healed enough to crawl about under it) is to drain the oil and drop the pan checking the condition of the oil (smell?) and running a magnet through the stuff for any metallic chips...and by feel for any non-metallic chips. Then replace the spin off oil filter and check all fittings and connections (although there was no indication of oil leakage anywhere after running the engine for a few minutes after the incident)

      Is there any allowable fore 'n aft play between the pulley and case? If not, and there is some play, what do I do then?? 

Carl.....   If an internal crankcase fire / explosion did occur, It points to the fuel delivery system having a problem somewhere.....   Could be the fuel mechanical fuel pump, fuel pressure, carb float level among other things....

The VW type 1 engine has no oil seal on the pulley end of the crankshaft normally....  It uses  the OLD labrynth seal and oil slinger technology of the '30s....  This is "wide ass open" to any air pressure differences....  A crank case fire could happen as you described and there may or may not be oil seepage from behind the pulley....

There will be some end play with the crank pulley....  All engines have it...  I can't remember how much, but its very small....  I doubt this will be a problem....

Get yourself well, then worry about the car.....   

Are you healing? I hope you're feeling better.

If the thrust is within spec (.003-.006?") then you'll just barely feel a little movement as you pull and push (for forward and back movement) on the pulley, but have you looked at the exhaust? Reconsider Stan's scenario; if fuel ignited in the exhaust, it would have the volume to hold enough smoke to come up through the pulley/tin gap for several seconds and is more likely to have happened. I don't think I've ever heard of fuel igniting in the crankcase before, although that doesn't mean it isn't possible. Have you gone through Leon's and Gordon's list?

I'm with Stan and Al.

It's probably not an intake issue, although it still wouldn't hurt to check your jets.

Most likely, it was a BIG backfire caused by you pumping the gas a few too many times, loading up the exhaust system with gas fumes and raw gas which ignited (probably when an exhaust valve opened) quickly burned, turning into a small explosion which didn't burn off all of the gas (a quick explosion consumes available oxygen, still leaving some fuel vapours) and the remaining turned to fuel-rich smoke.

The most open point of the cooling tins around the engine is at the pulley, right between the exhaust pipes.  You might have a loose exhaust gasket under there, or maybe your slip-connections on the exhaust pipes aren't super tight.  I have no clamps on two of my slip joints, simply because there isn't any room to fit a clamp.  That hasn't been a problem, but it is still a loose connection which could expel smoke if it backfired.  You could look for exhaust leak evidence at the clamps and flanges - black soot at the connection.

Still, I don't think that this is a major problem.

Carl also wrote:  

     The only explanation that makes sense to me is that the pressure of the explosion affected the integrity of the bolted connection between the pulley and the cam (allowing the smoke to escape) andaccounting for the subsequent ragged idle, vibration, and ratty noise.

Nope, I don't buy that.  There is no "bolted connection between the pulley and cam". The pulley is bolted to the crankshaft and drives the cam via a gear.  The only gap there is the un-sealed slinger washer behind the crank pulley.  The cam gear is buried inside of the case and has no "cover" behind the crank pulley.  All that's behind the pulley is the oil pump.  If the oil pump or connections were leaking, you would have to run the engine for over a minute to get the exhaust hot enough to burn off leaked oil and make smoke and the smoke might linger for 2-3 minutes after stopping the engine.

I still think exhaust backfire.

Yeah I was thinking "bolted connection between pulley and crank"

"cam" was a mistake I simply didn't catch...not dopey ignorance.

But I'm sticking with "crankcase conflagration, fire and brimstone"!!!

Simply too much billowing smoke rolling up for TOO LONG to be a the result of a backfire...plus I later slid under the engine with a drop light, inspecting everything, without any indication whatsoever of damage, fire, or soot.

Nor was there any powered soot inside the exhaust tips.  

Gordon, It wasn't "lingering smoke" drifting and hanging about for "two or three minutes"

It was rolling billowing smoke resembling an evening news video of a raging apartment fire with black choking smoke being pumped through an open window!

It was NOT a short lived exhaust backfire.  

Help me understand.

If there is a leak in the intake gasket and carb cleaner is sprayed around the carb base and because of an air leak in that gasket, isn't the mixture enrichened by sucking the volatile carb cleaner into the engine causing the enginge sound to change.  That's what happens with adding the carb cleaning into the engine this way---right?  (And no leak means no carb cleaner gets into the engine to change the mixture.)

So my question is, if indeed there is a leak in the intake gasket what sort or problem is caused by that ?   Or: What trouble is caused by a faulty manifold gasket?

Thanks!

Jack- Generally, an intake leak means a lean mixture on the affected cylinders. If serious enough (and left long enough) it can burn valves (and seats), and destroy pistons.

It's a diagnostic way to determine if the intake manifold gasket has been damaged (not always visible after a powerful backfire) or simply deteriorated, and needs replacement...the idle rpm will increase because the cleaner has been sucked into, and enriched, the air/fuel ratio. Unless the gasket is replaced that carb will be unadjustable, with a ragged idle. 

An easy elimination test to run before moving on to suspecting clogged jets.

Cool--ALb and Carl---thanks.  I understand --thanks for the explanation--I just didn't know.