A few 20/2 assembly questions

[Jens]

I am slowly putting Thumper back together and I will be asking some questions while I am doing this. Here are the first two:

I slipped the crankshaft back in today and I will need to adjust the tapered roller bearing preload. I am wondering what the best procedure is for doing that. All I can think of is reducing the gasket stacks symmetrically until the crankshaft feels too tight to rotate and then remove one thin gasket to make it feel 'right' .... doesn't seem too scientific to me. Any better methods ?

Sooner or later I will need to install the cylinders. The gasket thickness at the base of the cylinder needs to be set for correct piston clearance. The way that looks to me is to pick a number of gaskets, install them, install the cylinders, install the pistons/connecting rods, install the big end caps, install the heads, insert a small piece of soldering lead as a gauge, torque everything down and pray. If not right, take the whole friggin' thing apart again and start over ..... I am fine with most of that except for the concept of taking half the engine apart to try a different thickness gasket if the clearances are slightly off. Is there any way of doing this in a smarter way ?

Wish me luck with the camshaft timing .... I am not looking forward to that!

[Crofter]

I am slowly putting Thumper back together and I will be asking some questions while I am doing this. Here are the first two:

I slipped the crankshaft back in today and I will need to adjust the tapered roller bearing preload. I am wondering what the best procedure is for doing that. All I can think of is reducing the gasket stacks symmetrically until the crankshaft feels too tight to rotate and then remove one thin gasket to make it feel 'right' .... doesn't seem too scientific to me. Any better methods ?

Sooner or later I will need to install the cylinders. The gasket thickness at the base of the cylinder needs to be set for correct piston clearance. The way that looks to me is to pick a number of gaskets, install them, install the cylinders, install the pistons/connecting rods, install the big end caps, install the heads, insert a small piece of soldering lead as a gauge, torque everything down and pray. If not right, take the whole friggin' thing apart again and start over ..... I am fine with most of that except for the concept of taking half the engine apart to try a different thickness gasket if the clearances are slightly off. Is there any way of doing this in a smarter way ?

Wish me luck with the camshaft timing .... I am not looking forward to that!

Jens, You might be lucky and need similar thickness shim pack each side of crankcase but you do have to position the crank timing gear endwise to mate with the timing idler. You may wind up shuffling shims from one side to the other as they set both preload and position.  If you haven't already fit the con rod big end bearings and shims, I would do that first with out piston attached to the rod. It is also much easier to turn the engine over without ring drag or danger of disturbing the sleeve and O ring.

Same with the shim packs for setting the cydinder deck height, as you fear you will have to do some trial and error assemble and dis assemble and best not have to deal with the rings until final assembly.  You can assemble extra gasket thickness and conceivably tear some out without having to totally lift the jug off but you still have to lift it some. You may have to use some half gaskets too to tilt the cylinder to get the upper surface plumb so squish is even side to side. Perhaps you have had the cylinder face to bore accuracy checked and trued or get lucky so you dont have to use partial leaner gaskets. I chose to machine mine to get the bore straiight, the liner protrusion equal all around and the squish even.

Dont get in a hurry now that you have invested this much. You had a failure of one throw rod bearing so you should check to make sure it was not a cocked cylinder bore that contributed to that. If you find you have a large stack of gaskets you migh choose to use some metal ones to replace some of the paper as they squish less and retain height and torque better. you cant tear them out to discard like you can with paper though.

With practise you can make multiple moves together and have to do fewer backsteps but it is safer to make one move at a time and keep in mind which one has to be done first not to undo another dimension.

Maybe more confusion than help but perhaps some food for thought.

[Ronmar]

Crankshaft tapered bearing preload?  I used the cold endplay method.  Set a magnetic dial indicator base  onto the crank with the dial indicator/gauge plunger touching the block.  Try and push/pull the crank lengthwise and note the crank endplay.  Add or remove shim gaskets to tighten or loosen the bearing preload to get the crank endplay to about .005"-.007".  Without a dial indicator, that is just loose enough to just barely be able to perceive the "clunk" thru my bare hands as it shifts that .005"  As mentioned, make sure the crank to idler gear relationship is correct.

As for the squish, did you note the number and thickness of the installed gaskets, or any half gaskets during the teardown?  Add that same gasket thickness plus a few extra paper thin ones and assemble the cylinders/heads and measure squish. This should give you less squish than is desired.  You don't have to completely teardown, just remove the head and lift the cylinder enough to tear out the appropriate number of the extra paper thin gaskets and retest.  Say you want .045" of squish and you measure .065".  You need to remove .020" worth of gasket material to get to the desired squish If you tear out too many, you will have to remove the cylinder and add fresh gaskets.  It helps to use a micrometer or vernier caliper to measure the gasket thicknesses before you install them so you know how much each gasket removed will lower the cylinder/head down onto the piston at TDC.

If you have a large powerfull magnet, you can lift the cylinder with piston/rod installed by just removing the big end bearing nuts.  You use the large magnet with a crossbar/stick across the top of the piston/cylinder to keep the piston in the cylinder while you remove the whole assembly.  I use an old magnetron magnet for this, which is a fat horseshoe type that sticks to the piston head.  Thru the gap in the magnet, I slip a wooden stick that rests across the top of the cylinder.  This keeps the piston from falling out the bottom while I heft the whole assembly up and over the studs.  Since the magnet holds the piston near the top, I can rotate the crank to BDC after I remove the bigend cap, so there is no danger of the rod studs meeting the crank journal.  A short length of fuel hose across the ends of the studs is cheap insurance that the studs don't contact anything either. 

You can probably get an old magnetron from a marine electronics service shop for free.  They are normally discarded after their 2000-4000 hour service life. The Canadian Coast Guard used to have an electronics shop down at Malahat.  They probably change them out regularly like we do, so could probably rustle a dead one up for you pretty easilly.  Just remove the electronic bits and you have a large strong magnet that is pretty handy.  I have a bunch of them and use them all the time...

When you do your squish test, Use two pieces of lead out near the edges of the piston, one over each end of the wristpin(but not under a valve:) the same distance from the edge of the piston. A difference in squish from side to side across the piston will tell you if the piston is true in the bore/cylinder is perpendicular to the crank. If you have more than about .010 difference in squish across the top of the piston, you may need a half gasket under one side of the cylinder base to even things out to get the piston's line of travel parallel with the cylinder bore.  #7 shotgun shot is great for squish measurements.  The round uniform pellet has a very consistent squish characteristic when mashed.  I use a little dab of grease to keep them from rolling out of position.

[Ronmar]

I did some editing to my above post jens as I typed the original on an ipod.  I was pretty sure the spec was around .005", but I will revisit my notes as I was recalling that from several years back. 

About gaskets, you can tap out a new base gasket using the bottom of the upturned cylinder in a few minutes.  Just cut out the cylinder liner hole with scissors, place the gasket material onto the cylinder base, and tap out the stud holes using a small ball-peen hammer.  Tapping on the sharp edge of the holes with the rounded end of the hammer cuts thru the paper and cuts it pretty cleanly and in perfect line with the hole.  Once the first hole is cut, drop a bolt into it to hold the gasket in line and move to the next hole and drop a bolt into it.  once all holes are tapped out, tap around the edge of the cylinder to cut the outside shape of the gasket.  I tapped out all the replacement gaskets I needed for my 6/1 overhaul in an evening using the engines parts as templates, a small pair of scisors, razor blade and a small ball-peen hammer are all you need...

[Ronmar]

Jens
  Here is a link to the third part of Quinn Farnes excellent writeup on building up one of the Utterpower "kit" engines where he explaines the same bearing setup process.   It is about half way down the page. http://www.utterpower.com/kit_engines3.htm  He also covers some other things such as squish and half gaskets that you might find usefull.  I think there was a discussion about the cold crankshaft endplay figure back on the Lister Engine Forum(search preload?).  If I recall correctly, the basic jist is cold you want a little bit of clearance to allow for thermal expansion of the crank... 

[Crofter]

Jens, any procedures I have run across for tapered roller bearings call for some preload. Certainly there are considerations to be made for extreme temperature induced dimensional changes but that would be very small in your engine. The hammering from cyclic loads can be an issue as well as load induced distortion. The rollers basically sit on a ramp and the force of the power stroke will attempt to spread the crankcase housings and cause a sliding action from the tapers of the components; load applied will increase any clearance you see at rest.

If there is no preload or actually clearance, the unloaded rollers must be propelled around their orbits by the force of the cage rather than rolling contact with the races. They engage in skidding action and force realignment when they come down to being loaded at the bottom of their path. Preload keeps them in constant contact and in proper alignment with the races at all times even under the pulsed loads. It may seem counter intuitive but maximum contact pressures and squirming is potentially worse with the initially slack scenario. ~1000 psi combustion pressures on your size pistons is quite a poke to a bearing with a relaxed belly

[billswan]

Well guys I will vote with what crofter wrote I was about to write that preload on the tapered roller wins in my book almost every time. Crofter did a much better job of relating his thoughts than I ever could have on a written forum .

Billswan

[Ronmar]

FYI, Here is the discussion I was thinking about from over on the lister forum IRT TRB preload.  The .005"-.010" figure came from some indian listeroid manuals...

http://listerengine.com/smf/index.php?topic=4103.30

[potter]

Hee hee I know the feeling.

     Potter   

[Crofter]

My engine which was not factory pre run was set up with about 8 thou. of TRB preload; I made up another gasket on either side bearing carrier to get just a slight bit of preload as the nuts were finally tightened down. No way they were set up with any free play. I did however have to take 7 thou shim out of con rod cap as it would not even start to crush the plastigage. Maybe their assembly guy got the specs mixed up.

They are probably expected to assemble an engine from parts in about two man hours. I probably spent two weeks straightening things out to reasonably 1st world specs. Of course that was my free time and actually fun but I went into it expecting to find lots of things out of whack so no disappointment.

[Crofter]

Jens, the paper wrap sounds workable. I have used thin shim stock and also carefully wrapping with black electrical tape. The sharp edges of the keyway definitely can damage the seal lips. Filling the seal cavity with grease is good. The leather seals had the garter spring to tension the lip; some of the newer synthetics do not have them. Either way grease should have no negative aspects.

The plywood should work if you can get some kind of a sleave of pipe or such to whump it with. Striking it alternately from one side and the other often just makes them walk around; you have to push squarely on the full diameter.

[Ronmar]

The paper would probably work, but are you sure you will be able to get ALL of it out?  I would say just lube the new seal liberally with oil, and use the tip of your finger in the keyway to depress the seal lip a little to keep it from touching the edges of the keyway as you slide it on.

As for seating the seal, you are probably not going to be able to push it in.  As Crofter mentioned, alternatly tapping around the circumference may or may not be successfull, and I have seen the metal seal cups damaged/distorted in this fashon as the seal gets twisted.  I also would use a length of 2" ID pipe, just long enough to slip over the shaft and rest against the seal face.  If the edge of the pipe is cut smooth and flat, you probably would not need the plywood ring, providing the pipe wall thickness is great enough to cover most of the seal face.  The plywood spacer you mentioned would make the pipe diameter and wall thickness less critical.  Then I would hold a block of wood over the end of the pipe so I could tap it with a hammer right in the middle to drive the seal home evenly.  If everything is clean and lubed, and it is the correct seal for the application, and the force is applied evenly around the seal, it shouldn't take much force to tap it home.  A short length of thickwall PVC pipe might even work as a driver... 

[billswan]

I did the setup of the crankshaft end supports today and have roughly 3 thou free play but I have only torqued up 2 nuts on either end and suspect it will probably be closer to zero in the end (I will double check when tight). I figure that is a good compromise between end play and preload since this is essentially an unknown

Jens

I would shoot for as close to 0 end play as possible. Or even a low negative like .003 preload, the crank case is only flexed out 1.5 thousandts on each side. That will never hurt the bearings or any thing else even at full load and hot operating temps. At least that is how I see it.

Billswan

[Ronmar]

Jens, the paper wrap sounds workable. I have used thin shim stock

The shim stock is a great idea. That way I am sure there isn't left-over paper in there. Thanks !

I was just thinking a piece of 35MM film negative might also work well...

[sailawayrb]

Here's another discussion we had regarding TRB preload:

http://slowspeedengines.com/forum/showthread.php?t=770

There appears to be many opinions on this subject.