Removing Woodruff Key on 1995 2.2L Subaru legacy

[jimdanahand]

The "easy out" is a screw extractor. Even with the radiator removed I don't have enough room to get my drill in for a straight shot. I've picked up a right angle bit to drill the hole. Maybe I should wait and get a left handed drill instead of using the right handed bit to drill the pilot hole for the screw extractor. I also picked up some JB Weld but my son has tried it before and found it not to be that tough. I've got to be careful not to get too much on the end of the bolt or I'll end up "welding" the bolt into the hole. Does anyone else think JB Weld is worth a try before drilling a hole?

 

Jim

[jimdanahand]

Or did you use Loctite?

 

I didn't use Locktite fortunately so hopefully it should be easier to backout the bolt. But I'm still not convinced JB Weld will hold enough to back it out.

 

Jim

[OB99W]

Many screw extractors come with correctly-sized left-handed bits, sometimes cobalt type; for example, http://www.tools-plus.com/hanson-11119.html . If yours came with drill bits, see whether they're left-twist.

 

You didn't answer my question about the condition of the 1/4" of exposed thread in the crank; if damaged, extraction will be more difficult. Did the part of the bolt that broke off leave behind anything? You said that you didn't use loctite, but did you lubricate the threads? If not, do so before trying to remove the bolt remnant.

 

JB weld might work in the application that Gary mentioned, but probably not if the intent is to just glue something onto the broken end.

 

When drilling, be sure to center-punch beforehand so that the bit doesn't walk, and naturally try to stay on-axis.

 

By the way, what number (or mark) is on the bolt head?

[idosubaru]

don't bother with the JB weld, it'll be a waste of time.

 

DO NOT USE AN EZOUT. you break that off inside you're crank and you're really in for a world of hurt. please don't use it, yes it might work. but they are very, very brittle and break easily. seen it happen many times, and getting a stuck EZ OUT removed is a PITA. do not follow anecdotal advice on this one, just because it "has" and "can" work does not mean the risk is worth it. EZ outs are horrid, i personally think the things should be illegal. if the EZOUT works, the method below will work as well, i gaurantee it...and without the risk. ALSO...the twisted style EZ outs suck also because they make your job harder if they don't work. as you tigthen they force the remaining material "out" and make the bolt tighter inside the bore of the hole...making further efforts more difficult if it doesn't work.

 

spend the effort and time waiting for a left handed cobalt drill bit (or a couple in different sizes). that bolt isn't in there very tight, it might just back it out. any drill should have a "reverse" setting, you don't need a special drill to drill "left handed". all you need is a basic drill, a right angle attachment (or just a right angle drill) and the left handed drill bit. so long as the drill has a reverse on it...which all of mine do, never seen one that doesnt, you're golden. a right handed bit would work fine and that's the route i would go, i wouldn't wait, but being your first time dealing with this i'm trying to be cautious for you. as a last resort...you can always continue drilling out...drill one small hole, then one or two more sizes bigger until the bolt literally falls apart...that will always work. i would save time by doing one of the following below though.

 

i would drill a hole, then insert needle nose pliers into it...then "open" them while they're in the hole and see if they bite enough to turn the bolt. if that doesn't work i'd plan on threading a bolt inside the hole with some locktite. let it set and use a socket to turn it out. i think that's the easiest solution and won't be prone to causing trouble if something goes wrong. just don't use some high grade steel bolt that's difficult to drill or anything just in case.

 

here's the best method for keeping it centered: get some tube the same diameter as the crank shaft hole. drill a hole in the center of it for your drill bit. fit this in the crank bolt hole and drill the crank bolt through it....this will keep it centered.

or....use a dremel tool if there's room, to grind a flat area to keep center. or use a punch to create a center divot to start the drilling. or use a very tiny drill bit to start the hole.

[OB99W]

[...]or use a very tiny drill bit to start the hole.

Small bits are usually easier to start a hole with, but they're also easy to break if you apply pressure off-axis. Since they're made of hardened steel, you can wind up with a problem similar to a broken-off "easy out". Starting with a small bit is fine, just not "too" small; if you can deflect it with a little side pressure, there's a good chance of breaking it while drilling (using a hand-held drill in an awkward position, anyway).

[idosubaru]

Small bits are usually easier to start a hole with, but they're also easy to break if you apply pressure off-axis. Since they're made of hardened steel, you can wind up with a problem similar to a broken-off "easy out". Starting with a small bit is fine, just not "too" small; if you can deflect it with a little side pressure, there's a good chance of breaking it while drilling (using a hand-held drill in an awkward position, anyway).

 

good point, that means maybe you should just use a very small bit only to start the divot, not actually drill a hole of any depth.

 

i have another idea...as if you needed another..

i was wondering....if you drilled a hole all the way through the remaining crank bolt and tapped it ...then threaded a very long bolt into it...long enough to bottom out, and kept turning it after it was bottomed out, since it is only threaded into the remaining crankshaft bolt it will never tighten and it will keep rotating the crank bolt. the remaining crank bolt will either go further in the hole, or come out. i just can't picture which way the remaining crank bolt will go. but if it goes further in, then you'll just need to get a left handed threaded tap and bolt. i think this is the method i would try first. given that the bolt isn't actually torqued or seized in, i believe it'll come right out and you avoid any ez outs or the like. make sure it's not a little tiny hole and bolt, make it large enough to handle the job. i'd say 3/16" would be golden...but i don't have any to look out in front of me....

 

actually the left handed tap might actually draw it out, the only down side to this is that a right handed tap might "screw it in", you'd have to keep an eye on it. i'm betting the left handed tap would pull it right out, it's tight work threading a hole. use a large tap...a broken tap is about as suicidal as a broken ez out, but i've never broken one yet. be sure to use the correct size bit and work the tap properly and it shouldn't break, that's what they are made for.

[ron917]

the easiest way and what i would do is tack weld a bolt onto the sheared off end. let it cool down and back it out with the bolt you just welded on. this is by far the quickest and easiest method and will work every time here as the crank and bolt are easy to weld on.

 

Like grossgary said, that trick always works. But, I use a nut instead of a bolt. Hold the nut over the remains of the bolt with a pair of pliers, and feed your MIG or flux core wire into the hole.

 

This is an old trick, and works even with badly corroded fasteners (the heat helps break the corrosion). Your bolt is not corroded, so it should come right out. If the nut breaks off because your weld wasn't good, no big deal - just try again.

 

As others have said, stay away from the ez-outs! They are brittle, and harder than any drill bit. When they break, you can sometimes get the pieces out with the "weld a nut on it" trick, but if not, you need to spend hours with a Dremel and diamond bits. Or, hire a machinist with an EDM machine which will probably require removing the crank.

 

It sucks that the bolt broke. It had to be damaged or weakened from previous stress. It should not have broken with 125 lb-ft.

[OB99W]

[...]i was wondering....if you drilled a hole all the way through the remaining crank bolt and tapped it ...then threaded a very long bolt into it...long enough to bottom out, and kept turning it after it was bottomed out, since it is only threaded into the remaining crankshaft bolt it will never tighten and it will keep rotating the crank bolt. the remaining crank bolt will either go further in the hole, or come out. i just can't picture which way the remaining crank bolt will go. but if it goes further in, then you'll just need to get a left handed threaded tap and bolt. i think this is the method i would try first.

If the "long bolt" is right-hand threaded, I suspect if the bolt remnant is free to turn it will be driven further in by this approach, until there's bottoming, and then things will jam; if the remnant is stuck, the additional bolt will eventually bottom, then jam.

 

If a left-hand threaded bolt is used, it may help turn the remnant out. However, if the remnant is stuck, the additional bolt could possibly thread until it bottomed, and perhaps jam.

 

 

given that the bolt isn't actually torqued or seized in, i believe it'll come right out and you avoid any ez outs or the like.

But there's no guarantee that the remnant isn't seized. As I mentioned before, when a bolt snaps, it's exceeded its elastic limit. That means it's stretched (yield zone) and won't return to original dimensions, particularly in the area of the break. Depending on tolerances, there could now be an "interference fit" between the threads of the bolt remnant and the crank's threading.

 

Gary, I liked your previous idea of drilling, tapping, and using loctite on an unhardened bolt; if it breaks, at least there's just a soft piece of metal, no worse than the original remnant, to deal with (as long as the tap doesn't get broken; see below).

 

make sure it's not a little tiny hole and bolt, make it large enough to handle the job. i'd say 3/16" would be golden...but i don't have any to look out in front of me....

I don't have one at hand either, but I believe the crank bolt diameter is 14mm, so 3/16" would be "small" compared to that, but probably not too fragile.

 

 

actually the left handed tap might actually draw it out, the only down side to this is that a right handed tap might "screw it in", you'd have to keep an eye on it. i'm betting the left handed tap would pull it right out, it's tight work threading a hole.

Nobody is going to like me for this, but... taps have a taper, and sometimes jam and break, and they're also hardened. If the tap is being used to allow insertion of a bolt that will be used as removal tool, the tap needs to be properly lubricated, and backed out frequently to remove chips, or it may jam. If the tap is being used as a removal tool itself, then jamming might be "okay", but you run the risk of breakage.

 

Yeah, I'm a real "killjoy".

[OB99W]

Like grossgary said, that trick always works. But, I use a nut instead of a bolt. Hold the nut over the remains of the bolt with a pair of pliers, and feed your MIG or flux core wire into the hole.

The problem is, apparently the remnant is recessed by 1/4" or so into the crank.

 

 

This is an old trick, and works even with badly corroded fasteners (the heat helps break the corrosion). Your bolt is not corroded, so it should come right out. If the nut breaks off because your weld wasn't good, no big deal - just try again.

See my previous posts concerning bolt deformation once elastic limit is exceeded and the bolt goes through the yield zone.

 

 

As others have said, stay away from the ez-outs! They are brittle, and harder than any drill bit. When they break, you can sometimes get the pieces out with the "weld a nut on it" trick, but if not, you need to spend hours with a Dremel and diamond bits. Or, hire a machinist with an EDM machine which will probably require removing the crank.

Yes, breaking off any hardened tool will make things worse.

 

 

It sucks that the bolt broke. It had to be damaged or weakened from previous stress. It should not have broken with 125 lb-ft.

I'd still like to know what grade/class this particular bolt claims to be.

[ron917]

The problem is, apparently the remnant is recessed by 1/4" or so into the crank.

 

Oh, missed that bit of info. Cut a bolt short and bore hole through it, then weld as in the nut trick. A short piece of tubing followed by a nut welded on might work as well.

[idosubaru]

i mentioned welding a long time ago, he does not weld and wasn't interested in pursuing that option, i think it's time to move on unless he informs us otherwise. i would feel very confident of this bolt coming out based on the observations made thus far. i've had very good success getting out corroded, seized and very difficult bolts, i don't think this one, not even being under load or seized will be all that bad. i had materials engineering course and understand yield very well and agree that won't help, but i think it's not a huge obstacle if you proceed slowly and pay attention.

 

and good points about the tap...read up or ask about using one if you're not familiar with using them. lubricate, allow it to cool, and thread, reverse, thread reverse to break up chunks as you go. remove to clean at intervals.

[OB99W]

[...]i would feel very confident of this bolt coming out based on the observations made thus far. i've had very good success getting out corroded, seized and very difficult bolts, i don't think this one, not even being under load or seized will be all that bad. i had materials engineering course and understand yield very well and agree that won't help, but i think it's not a huge obstacle if you proceed slowly and pay attention. [...]

I agree, even if the bolt remnant proves somewhat reluctant to come out, it should be doable. Lubrication and slow, careful work are the keys. Mostly, if anything seems to be seizing during the extraction, it's time to take a break and rethink the situation.

[Skip]

As I recently became a member of

"The hot dam crank pulley bolt" club

I thought I throw a couple pennies and a pict in the pot.

 

Here is a picture of the old and the newer pullies

and the tale of two crank bolts, why two?

 

Look close the one is 0.32" longer than the one that came loose.

The longer one is from an EJ25, the shorter from the

engine in question an EJ22T.

 

The bolt in mine came loose and well we all know the result.

I believe I am going for the Rallitech twin dowl rod/pin solution.

 

The dia. of the bolt shank is 0.472" (12mm)

No class marks on the heads but the longer bolt has

a "H" cast into it.

 

I say get a left handed tap (MSC.com has them.) and a left handed bolt.

Using a right hand tap could just drive the fastener deeper

into the crank.

I also like the step drill op. with a left handed bit.

 

Hope this adds some fuel to the fire.

[jimdanahand]

Okay, I tried the JB Weld and as most suspected it failed. I had bought the Easy Outs but after seeing all the comments I've elected to go buy a few left turning drill bits and try to back it out while progressively drilling a larger hole. I was planning to use a punch to get a pilot hole to avoid the bit "walking", but in addition I like the idea of using a tube to help keep me centered and protect the threads. Not sure what I'll use as a tube just yet. Because the bolt is recessed inside the crank, I'm going to pass on welding (even though I found a friend with a welder). I've never tapped a thread so I'll investigate further if the bolt doesn't back out while I drill counter-clockwise or after I get my hole and stick the needle nose in the hole to try to turn it. Wish me luck! Thanks for all the advice.

 

Jim

 

P.S. - There are no identification markings on the broken crank bolt

P.S.S. - Spoke with a local Suby expert who says he had never heard of breaking the crankshaft bolt. His guys usually use a impact wrench to tighten!!

[jimdanahand]

Look close the one is 0.32" longer than the one that came loose.

The longer one is from an EJ25, the shorter from the

engine in question an EJ22T.

 

It looks like my broken bolt is longer (judging from the pic). Just to confirm, my threads start 1-3/8" down from the head. If this is the same as yours, they may have used an EJ25 bolt in my EJ22 engine. Not sure if this means my bolt bottomed out and the remnent will be highly torqued! If so, I may not be able to back it out.

[OB99W]

[...]I was planning to use a punch to get a pilot hole to avoid the bit "walking", but in addition I like the idea of using a tube to help keep me centered and protect the threads. Not sure what I'll use as a tube just yet.

Yes, a tube or other guide can help both with centering and being on-axis, which is just as important. I've used coupling nuts (see http://www.chicagohardware.com/nuts.htm if anyone doesn't know that term), pieces of water line, brake line, etc., sometimes alone but often nested in each other. A trip to hardware and plumbing supply places can help your imagination. (By the way, the coupling nuts can be useful both when the break is recessed and when the remnant projects beyond the surface. For the recessed situation, obviously the hex has to fit inside the threaded hole, so it can be a little tricky to find the correct size. For the times when a portion is above the surface, matching threads can help.)

 

 

Because the bolt is recessed inside the crank, I'm going to pass on welding (even though I found a friend with a welder). I've never tapped a thread so I'll investigate further if the bolt doesn't back out while I drill counter-clockwise or after I get my hole and stick the needle nose in the hole to try to turn it.

Should it have to go that far, there are a few things to be sure of when tapping:

 

1) Drill the correct diameter hole for the particular tap; too large, and the threads won't be full-depth and will be weak -- too small, and the tap may jam

2) Use appropriate lubricant; don't, and the threads will suffer and the tap may jam

3) Start the tap on-axis; if it's cocked, it will likely jam.

4) Back the tap out frequently and clear chips; don't, and the threads can become ragged or the tap may jam

5) Take your time and trust your instincts; if it seems that too much effort is needed, stop and determine why, or the tap may jam

6) I don't mean to alarm about the jamming , but ignoring things like the above are typical ways that some beginners have trouble

 

 

Wish me luck! Thanks for all the advice.

You're welcome. I already wished you luck back in post #24 of this thread, but I'll gladly wish you some more of the best kind.

 

 

P.S. - There are no identification markings on the broken crank bolt

hmmm

 

 

P.S.S. - Spoke with a local Suby expert who says he had never heard of breaking the crankshaft bolt. His guys usually use a impact wrench to tighten!!

Breaking that bolt is certainly not a common problem.

 

 

Look close the one is 0.32" longer than the one that came loose.

The longer one is from an EJ25, the shorter from the

engine in question an EJ22T.

 

It looks like my broken bolt is longer (judging from the pic). Just to confirm, my threads start 1-3/8" down from the head. If this is the same as yours, they may have used an EJ25 bolt in my EJ22 engine. Not sure if this means my bolt bottomed out and the remnent will be highly torqued! If so, I may not be able to back it out.

Skip, thanks for the "heads up".

 

Jim, I certainly hope that someone didn't previously use a too-long bolt, and this all works out for you.

[idosubaru]

if you known someone with a welder, that would be time well spent to ask him to have a look and find out what kind of experience he has. this is a very easy welding job...it doesn't have to look good, it's not a finished product, it just needs to come out. with a welder this job becomes a very easy 2 minute fix. i've done it numerous time, weld a bolt or nut to the end and then use a socket to get it out. you could literally be done in 2 minutes. that certainly beats spending all that time centering the bit, grinding, tubing, drilling, drillling again, tapping, bolting, locktite...etc, etc.... if it were me, i would go this route, but i have a welder and have done it before. it is the single best method for removing a stuck fastener.

 

good luck, look forward to hearing good news.

[Skip]

My question here is

Why did the FHI engineers

only allow 0.385" of crank shaft

(0.22" of key)

stick out into the bore of the pulley

(the deapth of the hub is 0.55")

 

This damper weighs in excess of three lbs

and turns the AC comp, PS pump and the alt.

A bit of torque is required to turn these units.

 

IMHO

Welding in that small hole will make keeping the weld

"splatter" from welding the broken bolt to the crank

very difficult.

As Jim says it's broken off in the hole.

 

My MIG torch tip will cover the entire hole allowing no vision of the weld puddle.

This is out of the car, even with the rad removed,

getting in there in the car would be a challenge.

 

BTW

I did thread the longer bolt into the EJ22T crank

with out the pulley - it did not bottom out

before the bolt shoulder hit the crank.

 

EJ25 pictured as it is on the stand, EJ22T is the same

.

[idosubaru]

it does suck, but i think a wire feed is worth looking into. i've done some tight space welding with a wire feed, the thin wire feeds through very tight spaces without getting the tip close to the object. not ideal, but worth a look i think. almost positive my small wire feed would fit this job.

[jimdanahand]

BTW

I did thread the longer bolt into the EJ22T crank

with out the pulley - it did not bottom out

before the bolt shoulder hit the crank.

 

Skip - thanks for checking. Great news that the longer bolt didn't bottom out. That gives me more confidence that I might be able to back it out.

[Skip]

Jim, You are most welcome.

As we all have said "here's good on ya"

 

 

Gary, I don't doubt your welding skills, I to have

a small MIG set up and a larger one.

(a.k.a. wire feed)

 

Even with the small setup,

getting your hood (so you can see) and torch

into a position where a proper weld could be made

without thread gauling splatter would be a job for someone with

a lot of experience.

A TIG set up may be a better.

 

The right angle reversable drill and left handed bits and taps

Would be my choice.

 

The chart for selecting the proper drill size for a tap

in most cases is set up for 75% thread depth.

In a case like this where breaking a tap would result in dire consequences I would go about 0.020" bigger.

 

If you choose the left handed tap/bolt method.

 

Do not run the tap through the remaining broken bolt.

As OB99w says the taps are tapered

(if it's a gun tap not a bottom tap).

This will result in the left handed bolt locking in the remains and giving you purchase to back out the broken piece.

 

Please use a sulfur based cutting oil.

This is sold in most plumbing dept. as it is used for threading steel pipe (as in gas lines).

[NOMAD327]

I’ll make a non relevant comment here about Loctite for people subscribing to the thread. All Loctite products break down at 200 degrees, even the super strength red stuff. This is useful in that when it is used, the threads are shielded from corrosion, and loosening from vibration. When the time comes to disassemble, heat is frequently useful, and will also break down the Loctite bond.

 

I would try one more simple thing on this job, and since we are planning to weld to the broken face of the bolt, I don’t see any harm in this, but you all can correct me. I would obtain a very small cape type cold chisel, (slanted diamond shape tip), and try tapping the bolt out. The failure described in this thread should have caused no thread deformation, and I can’t believe the bolt is stuck that thoroughly.

 

I would heat to about 200 degrees, then quench the threads with oil, this is not hot enough to harden the metal, but should wick the oil into the threads. Then pick two opposite points on the face of the broken off bolt face. You don’t want to do anything that would mushroom the end or make it fit tighter into the bore, but slight impacts at say 3 o’clock, and 9 o’clock directed at an angle like pushing a broom, and in counter clockwise direction may spin the thing right out. Not too close to the OD for fear of distorting the bolt diameter, but not too close to center, or there’s no leverage. It’s easy to try, not expensive to set up, and if it doesn’t work, you just have to flush away the oil with a spray solvent and you can then do the welding fix. The one thing that would mess this technique up, is if the shank diameter of the bolt is less than the thread diameter. In that case, the bolt will tend to rock left and right in the unsupported diameter, instead of rotate. I think it would still be worth a try in any case.

[idosubaru]

Gary, I don't doubt your welding skills

i'm not very good at welding, just discussing.

[Skip]

Just so we all have a better "mind eye" view

of what Jim is up against,

 

here is a shot from the work position.

 

I placed a piece of yellow chalk into the hole

to simulate the broken bolt.

 

Now he will probably remove the rad and fans but the AC

condenser coils are still there.

 

[jimdanahand]

I would try one more simple thing on this job, and since we are planning to weld to the broken face of the bolt, I don’t see any harm in this, but you all can correct me. I would obtain a very small cape type cold chisel, (slanted diamond shape tip), and try tapping the bolt out. The failure described in this thread should have caused no thread deformation, and I can’t believe the bolt is stuck that thoroughly.

 

I would heat to about 200 degrees, then quench the threads with oil, this is not hot enough to harden the metal, but should wick the oil into the threads. Then pick two opposite points on the face of the broken off bolt face. You don’t want to do anything that would mushroom the end or make it fit tighter into the bore, but slight impacts at say 3 o’clock, and 9 o’clock directed at an angle like pushing a broom, and in counter clockwise direction may spin the thing right out. Not too close to the OD for fear of distorting the bolt diameter, but not too close to center, or there’s no leverage. It’s easy to try, not expensive to set up, and if it doesn’t work, you just have to flush away the oil with a spray solvent and you can then do the welding fix. The one thing that would mess this technique up, is if the shank diameter of the bolt is less than the thread diameter. In that case, the bolt will tend to rock left and right in the unsupported diameter, instead of rotate. I think it would still be worth a try in any case.

 

I'm not sure I'm comfortable using a torch to heat the crankshaft. I assume it's safe or you wouldn't suggest it? I'm not clear on how I use the chisel. Since I'm planning to use a left turn drillbit. Will using the chisel mess up my ability to get a good center punch for a starter hole? How am I using the chisel? I think the broken bolt is too far recessed to get the angle you suggest to help back the bolt out by hitting it at an angle. I still don't think welding is feasible given the small diameter and how far recessed the bolt is.

 

I took a break from this job to go catch a King Salmon on the Deshka river! I'll probably start back up on Saturday.