This Cavalier belongs to a young woman with very little savvy when it comes to properly maintaining her vehicle.  Well, she has more of an understanding now than she did before; what she discovered is that it’s not a good idea to ignore an overheating problem, even if the car does seem to be running well most of the time. 

When the Cavalier came to us on the hook, the water pump had been replaced, but the engine was hydro-locked when we tried to spin it over.  Removing the spark plugs revealed a #2 cylinder filled to the brim with liquid.  Bumping the engine (it was out on the parking lot at the time) blew the coolant out, but adding coolant to the radiator simply refilled the cylinder and pushed more out the spark plug hole.  This 2.2 had been overheated to the point that simply replacing the head gasket would only serve to be a waste of time. Exploratory surgery later revealed head warpage in the 0.010 inch range, and the block was warped about .004.  Even if the head hadn’t been warped, it was likely that the rings had lost their tension.

     Not all engines are subject to fail when they overheat.  I worked on a 1985 Tempo about ten years ago that had been run with plain water in the cooling system for so long that the radiator and engine block were saturated with muddy crud.  The owner kept seeing a red “Engine” light.  Fords of that vintage with red Engine indicator light would illuminate the light if either the oil pressure or the engine coolant drifted into the danger zone.  Well, she had taken it to a shop that was accustomed to seeing “Check Engine” lights on GM cars and since they weren’t familiar with the red “Engine” light on Fords, they sent her on her way.  Depending on how far she drove, the Tempo’s “Engine” light would illuminate and the car would stop running once or twice a day. She would pull over and wait awhile for the engine to cool, and then fire it up and drive some more.  This had been going on for a week or two.

            Removing the thermostat in that 2.3L I found that it cylinder head had been hot enough to turn it gun metal blue.  Thinking I had a basket case on my hands, at the customer’s request, I removed the radiator and sent it to have it cleaned and did a cooling system flush along with a thermostat replacement and a 50/50 mix of clean green coolant and water.  That Tempo didn’t smoke or use oil and it didn’t blow a gasket.  In spite of the cast iron cylinder head and structurally bulletproof design of the 2.3L HSC 4 cylinder engine, that car was a miracle on wheels.  I knew why that one overheated.

Back to the Cavalier now, the local salvage yard brought me an engine for $550.  Interestingly, Johnny the salvage yard guy told me he can strip one of these 2.2L engines down to a long block and sell it for $750 to somebody who needs an engine for an S10, but in these parts, $550 is about all a 2.2 will bring.

            Victor is right out of high school, but he’s one of my most careful and conscientious students.  He prefers to work alone, and even when he’s working on a trainer vehicle, he majors on minors to the point of pedantry; every hole that’s supposed to have a bolt in it gets one, even if the hole has to be drilled, tapped, and helicoiled.  He works a little slower than some of the other students and asks a lot of questions, but his finished work is beyond reproach.

            Victor disconnected everything, dropped the H frame, and lowered the whole powertrain assembly out the bottom.  The replacement engine got a rear main engine oil seal and the 3T40E got a torque converter seal while we were there.  He did an inspection of the brackets and stuff on both engines and made the parts changes he deemed necessary, then recoupled the transaxle to the engine and raised the powertrain back up into the car with the powertrain lift.  The new water pump from the original engine was installed on the replacement engine along with a new gasket and a new thermostat. He already had the fuel lines and most of the wires connected when he realized that the oil pressure sending unit was different (too many pins) as was the crank sensor, which was a variable reluctance ($10, two wire) sensor on the 1997 and the one on the engine he was installing (which, it turned out, was a 1996 model) was a Hall Effect ($100, 3 wire) sensor. 

This was a pretty intense study of an engine neither of us had been familiar with before.  The Chevy techs probably learned all this in an Atlanta classroom about decade ago.

One question that arose was

“Why are the two sensors different?”  Certainly there were other significant changes besides the number of connector pins!   We had figured this was a plug-and-play deal, but it was rapidly turning nasty.

A little research revealed that the 1996 2.2L Cavalier Hall Effect crank sensor feeds its signal directly to the PCM, which in turn fires the twin coil packs via module control wires A and B.  On the later engine, the 1997 2.2L VR sensor feeds its signal directly to the ignition module, which fires the coils on its own (with PCM spark timing input).  But the ignition module on the 1997 model is a bit smarter and generates the 7 pulse CKP signal, sending it to the PCM for injector timing and rpm input.

So what it amounts to is that the 1996 and 1997 engines had a different ignition module, wire harness, AND a different PCM.  Interesting, but not surprising. The 1996 module wouldn’t have worked in the 1997 model car anyway. The second question Victor and I had to consider was this:

“Are the crank sensor trigger wheels the same on both cars?” 

The answer to that question was a resounding yes. A little research revealed that the crankshaft was the same on both vehicles. It had to be the same because the part number was the same.  With that in mind, Victor removed the 3 wire sensor from the 1996 engine and walked over to the engine shop (it’s in a different building) where he removed the plastic 2 wire 1997 crank sensor from the original engine.  He installed the brass 3 wire crank sensor in the junked 1997 engine and it slid in there just fine.  Of course, the tip of the 1996 sensor was a bit smaller, but from the sensor mounting bracket to the end of the sensor was the same, and the o-ring seal was the same size, so there wouldn’t be a clearance problem from the tip of the sensor to the trigger wheel.  Now to put the 1997 sensor in the 1996 engine.  It wouldn’t slide all the way in.  Now what?

     Speaking of Tempos and salvage parts, we replaced the transaxle in a 1994 Tempo with a 1990 salvage yard unit only to find that the 1994 halfshaft assemblies wouldn’t fit into the 1990 transaxle.  For some reason Ford had made the 1994 halfshafts bigger.  That wasn’t hard to fix; all we needed were the halfshafts from the 90 model wreck, because the outer CV joints were the same.

            In the case of the Cavalier, however, the 1997 sensor just wouldn’t go in the 1996 engine block.  We lubed, twisted, pushed, felt the innards of the hole, and everything else we could think of.  It really should have gone in. If the obstacle was a layer of varnish, it was too thin to feel with the naked finger and we just couldn’t see it with that final drive in the way.  A boroscope might have helped, but we didn’t have one.  I tried to do a part number search on the engine block, but there was no part number available for a bare block, and it would seem that a short block part number might be inconclusive.

A call to the salvage yard just muddied the water.  The interchange book said the 96 engine had a 2 wire sensor and the 97 engine had a 3 wire sensor.  That was backwards, and we noted with some interest that the 97 sensor slid into the 96 engine block painlessly.  Was it possible that the swap would work in one direction but not the other?  If the holes were slightly different in size it would seem that way. 

The salvage yard guy said he’d never run into this problem before, but would bring us a 1997 engine anyway, and strip the first engine for an S10 long block sale.  Victor removed the powertrain assembly again (if he had done the sensor swap while the engine was still on the floor none of this extra work would have been necessary) and got it on the floor about the time the second engine arrived. A close inspection of the 96 block’s crank sensor hole revealed a port that should have accepted the 1997 sensor, and it even felt smooth, but try as we might we couldn’t get it to go in there.  The salvage yard guy couldn’t get it in either.  We reinstalled the old sensor and he presented us with a 1997 engine exactly like the one we had removed originally.

In a real world shop situation, somebody would have lost time and money on that deal.  We don’t charge labor, so it was primarily a rich learning experience for Victor.

Another rear main engine oil seal had to be ordered, because snatching the new one we had just installed in the first replacement engine wasn’t feasible.  The water pump was removed from the first replacement and installed on the second replacement engine with another new gasket.

Victor got the second engine installed in record time (for him) and a day later we started the Cavalier, got the air burped out, checked the cooling fan for proper operation and drove the car quite a bit to make sure it was okay.  The water pump had been replaced on the previous engine, too late, it seemed, and that appears to have been the reason for the original overheating concern.  We replaced a Toyota Camry engine for the same reason a couple of years back.

     If we had run a wire brush or a small brake cylinder hone into that hole, the sensor would have probably worked just fine, but we didn’t know that until Victor had removed the engine.  If I encounter a situation again, I will certainly check to make sure everything fits BEFORE the engine is installed.  Victor certainly will too!

For a tech in training, this little bit of learning was priceless.  That’s what preparing technicians for life in the real world is all about.