I'm asking this question in reference to my '98 JDM Subaru Impreza STi, but it would/could apply to any engine with rubber drive belts.

This is what a Subaru crankshaft position sensor looks like... when in one piece.



This is what the crankshaft position sensor of my EJ20K looks like today. I could spin the sensor in the hole, but there was no way that freakin' thing was going to pull out of the hole.


Yep... I'm suspecting explosives may be required to dislodge mine.

There's a chance that any fluids/oils used in the process of trying to loosen this stuck/broken crank position sensor might eventually drip down onto the timing belt. So... I've got a question about the use of penetrating oils.

Are penetrating oils harmful to a timing belt?

Here are a couple of different images I've found online that show what's below the crankshaft position sensor in this engine.






And if anyone has any advice on how best to try and remove this broken sensor, please chime in!

[This message has been edited by Patrick (edited 04-09-2018).]

I believe you solved your problem in that last pic. Simply remove the cover and start chipping away. Another way would be to have a volunteer hold a shop vac hose near opening to prevent pieces from falling into engine and start chipping away with a fine chisel and hammer. Careful not to damage the bore.

Still may be a challenge. May have dissimilar metals involved and if they've bonded together your into a more difficult job. Boring / honing and sleeving.

Spoon

------------------
"Kilgore Trout once wrote a short story which was a dialogue between two pieces of yeast. They were discussing the possible purposes of life as they ate sugar and suffocated in their own excrement. Because of their limited intelligence, they never came close to guessing that they were making champagne." - Kurt Vonnegut

quote Originally posted by Spoon: I believe you solved your problem in that last pic. Simply remove the cover...

"Simply"? Man, a whole lotta stuff needs to come off to get to that stage. I'm hoping not to need to go that route.

quote Originally posted by Spoon: May have dissimilar metals involved and if they've bonded together your into a more difficult job.

That's probably at least part of the problem. The sensor shell is steel and it's located inside of a bore in an aluminum block. I suspect some level of corrosion has occurred. The other issue is that crap builds up over time on the end of the head of the sensor that protrudes down towards the notches of the reluctor ring and the cogs of the belt drive. This then can sometimes create quite an issue when trying to pull it back out of the hole.

I'm confident I can eventually get that broken sensor up and out of the hole (one way or another!), but my main question still stands...

Are penetrating oils harmful to a timing belt?

[This message has been edited by Patrick (edited 04-09-2018).]

quote Originally posted by Patrick: but my main question still stands... Are penetrating oils harmful to a timing belt?

Probably, but it's unlikely that you will get any large amount on the belt. Can you get something screwed into the center of the sensor to pull it out should it come loose? If you can, shoot it with some R134a for about 30 seconds and try to pull it out. Maybe lightly bump the side of the timing cover as you pull upward.

quote Originally posted by fierofool: Probably, but it's unlikely that you will get any large amount on the belt.

If it wasn't a timing belt on an interference type DOHC engine, I wouldn't worry about it, but...

quote Originally posted by fierofool: Can you get something screwed into the center of the sensor to pull it out should it come loose? If you can, shoot it with some R134a for about 30 seconds and try to pull it out.

The problem is, even though I got the outer shell of the sensor to spin in the bore, it still wouldn't pull up... and I was able to pull pretty hard before the top (and most of the guts) of the sensor came off. There's either a lot of crap stuck to the bottom of the head of the sensor, or the bottom of the sensor has been "mushroomed"... similar to the one on the left in the image below.

I wouldn't worry about it, you're not submerging the belt.

Plus, if an engine component can't handle a bit of oil, that's quite shoddy design!

Even if a part is not expected to come into contact with a specific chemical in service, it's standard practice at my employer to choose materials that are "resistant" (in standardized tests) to most automotive fluids; gasoline, oil, grease, coolant, even brake fluid IIRC. Take this with a grain of salt, because I can't speak for the automotive industry as a whole.

I would definitely use anti-seize on reassembly!

I recommend a wood chisel that just fits the bore and strikes the sleeve. Tap it straight down the bore with just enough force to split the sleeve, then pull the pieces of the sleeve out with a needle nose and a bit of sardine can twisting. This would avoid any use of p-oil.
If you can take precautions to prevent the chisel or the sleeve from traveling out of the bore and contacting that reluctor wheel, might save more drama. If the chisel scores the sides of the bore, those marks should be blended out with crocus cloth before replacing the sensor. Aluminum creates a stress riser wherever surface damage coexists with thermal transients, which is any part of the block that heats up with combustion.

[This message has been edited by CJB118 (edited 04-09-2018).]

A pilot hole, a good screw, and a slide hammer? I guess it depends on the material of the center. One way or another, anticipate bits and pieces of the sensor outer body to be a bigger threat than oils.

quote Originally posted by pmbrunelle: I wouldn't worry about it, you're not submerging the belt.

For sure. It's possible no penetrating oil at all has seeped by the shell of the sensor, and if it has, it's only going to be maybe a drop.

quote Originally posted by pmbrunelle: I would definitely use anti-seize on reassembly!

Yes, I'm a great believer in using anti-seize... especially when dissimilar metals are in contact with each other.

quote Originally posted by CJB118: I recommend a wood chisel that just fits the bore and strikes the sleeve. Tap it straight down the bore with just enough force to split the sleeve, then pull the pieces of the sleeve out with a needle nose and a bit of sardine can twisting. This would avoid any use of p-oil.

I'll first use a drill to break up and remove (with a shop-vac) the rest of the brittle plastic "guts" and then yes, what you suggest sounds good.

quote Originally posted by viperine: A pilot hole, a good screw, and a slide hammer?

The problem is, there's about zero clearance between the bottom of the sensor and the high points of the reluctor ring. I wouldn't want to chance drilling through the bottom and/or putting a screw through the bottom of the sensor. I don't think the brittle plastic inside the shell would hold a screw while being pulled/pounded/pried up, but heck, I can certainly try it before I break it up to remove it.

quote Originally posted by viperine: One way or another, anticipate bits and pieces of the sensor outer body to be a bigger threat than oils.

Because the outer metal shell of the sensor is shaped like a small can, I believe that I'll be able to break up and suck out the insides without anything being able to fall through. Then it's "just" a matter of collapsing the outer shell and pulling it up... in theory!

Thanks for all the feedback everyone. I'll report any progress as it occurs.

Patrick, once you've pulled all the guts out, you should be able to see if the shell is deformed in any way at the bottom. You may be able to drill a very small hole in the bottom using a magnetized drill bit. Pulling upward with a seal puller should pull the outer perimeters in a bit. I've used a seal puller to remove things other than seals.

Patrick,

I used to be a Subaru tech... If you already have the front cover off, the rad out, etc, I would go ahead and change the belt. You're already there, new belt is 40$, and you don't *need* to replace anything else, though I would do the water pump just in case, and cam seals, as they tend to leak.

Re use the tensioner, you can put it in a vice and re use it. IIRC use a cotter pin to hold it after compressing. Watch out for the flat washer behind the tensioner too, don't lose it!!!

As for the sensor, I don't recall ever running into one of those ceasing into the block like that... It looks like it's surrounded by Fujibond (Subaru silicone) I would maybe try to drill a hole in the center and get it out with an ez out bit... Maybe give it some heat with a propane torch.

------------------
"Discord"
Red 1988 GT under restoration!

Let's Go Mets!

( ͡° ͜ʖ ͡°)

[This message has been edited by Shho13 (edited 04-09-2018).]

quote Originally posted by Patrick: Are penetrating oils harmful to drive belts? (V and Serpent) Are penetrating oils harmful to a timing belt?

Yes to both and not just "penetrating oils." Engine oils and most others hate all belts.
Timing gears often won't slip but teeth can fail in a few days to months later w/o warning and if valves hit pistons then expect to need another motor. Interference engines common to Subaru and others 1 or more Valves can bend or break and trash piston(s) Cylinder(s) and/or head(s).

You should clean all "gears" and idlers before you install new belt.

Some cam "gears" have tools to save time to install. Holds them so timing is good w/o trying to rotate everything to matching timing marks and/or Cam "gears" can't move from valve spring loads.
Have a ok belt now so you should rotate til marks match but find out what extra tools may help or is required.

------------------
Dr. Ian Malcolm: Yeah, but your scientists were so preoccupied with whether or not they could, they didn't stop to think if they should.
(Jurassic Park)

The Ogre's Fiero Cave

There are penetrating oils that are said to be safe for rubber and plastic but they aren't common, and they would be slippery so you wouldn't want them on a belt.

quote Originally posted by 2.5: There are penetrating oils that are said to be safe for rubber and plastic but they aren't common, and they would be slippery so you wouldn't want them on a belt.

yes.
Timing belt Teeth aren't only points that load to move the "gears." Teeth make sure the belt does not slip at all.
Nearly entire belt surface take loads for operating whatever but any oil/grease in the parts then only the teeth tries to handle all force to fight slippage. Worse oil can allow the belt teeth to try to climb up/out of the "gear's" teeth and make the belt too tight etc.
Most times is the small crank "gear" that strips just a few teeth. If have interfering valve train then bad news when just trying to start when cold and belt dies... happens on the highway and engine can easily self destruct.

SUCCESS !!!

I'll post details and images tonight of the technique I used... and what I found down the hole.

Im thinking either bend in a side and grab with pliers, or puncture or drill a small hole in the bottom of the tin bucket and then put a tool down thru the hole than has a bend in it a pick maybe, so you can catch the bottom of the tin bucket part and pull it out?

[This message has been edited by 2.5 (edited 04-10-2018).]

quote Originally posted by fierofool: You may be able to drill a very small hole in the bottom using a magnetized drill bit...

The idea of drilling down towards the reluctor wheel scares me.

quote Originally posted by Shho13: If you already have the front cover off, the rad out, etc...

I don't... and I want to keep it that way!

quote Originally posted by theogre: Engine oils and most others hate all belts. Timing gears often won't slip but teeth can fail in a few days to months later w/o warning and if valves hit pistons then expect to need another motor. Interference engines common to Subaru and others 1 or more Valves can bend or break and trash piston(s) Cylinder(s) and/or head(s).

Trust the Ogre to try and ruin my day.

quote Originally posted by Patrick: SUCCESS !!! I'll post details and images tonight of the technique I used... and what I found down the hole.

Coming up shortly!

quote Originally posted by 2.5: Im thinking either bend in a side and grab with pliers...

You're very close.

Okay, here's the goods!

First step - I used about a 1/8" drill bit to drill out and break up all the plastic innards (and sucked it all up with a shop-vac). Who knew that there was about a mile of fine copper winding in there! In this image, you can see the now empty sensor shell. It is very thin steel... and it was very tight in the hole. This gave me hope that no penetrating oil had gotten past it.




I used a pocket knife blade to get between the sensor and the wall of the bore in three places. Once the gap was wide enough, I used a small narrow flat-tipped screwdriver to bend the sensor shell in a bit more... and then I gently tapped the screwdriver down to the bottom of the outside of the sensor. I should also mention that this screwdriver has been rattling around in my tool box for 45 years and is very smooth. That's important to mention, as the rounded edges did not dig into the aluminum bore.




So I then grabbed on with needle-nose pliers, and with a bit of a twist and a pull, the sensor shell came up and out rather easily. In this image you can see the corrosion on the walls of the bore which had contributed to holding the sensor captive.




And the best news of all... there was absolutely no evidence that any penetrating oil had made its way past the stuck sensor and down onto the timing belt.

In this image, the edge of the belt can be seen (black with green lettering). The white crap is just powered corrosion that was displaced when the sensor shell was removed. I attached a short length of fuel line to my shop-vac, which I then stuck down the hole and vacuumed all that powdery crap out of there.




I am rather pleased how it all eventually worked out. I'm hoping that this write-up might help someone in the future deal with a stuck crankshaft position sensor in any vehicle that might have one. You can rest assured that I'll be using a bit of anti-seize on the replacement sensor.

Interesting to note that the hole the camshaft position sensor goes into on this engine is quite a bit larger than the sensor itself. No chance of it getting stuck in there.

Thank-you to everyone who offered advice!