Crankshaft Run-out

[303epps]

I'm currently measuring my parts for 2007 build. Have a 4340 crank that's been on the shelf for a couple years that I was measuring the journals and run-out on. Journals checkout fine but I have the following measurements for run-out.
#1- 0.0015
#2- 0.0012
#3- didn't measure yet.
#4- 0.0012

Bentley states maximum limit is 0.012. Options are to run it as it is, get the crank ground or replaced. Regrinding and balancing will be about half the cost of a new crank.

I would appreciate getting comments / recommendations on best approach.

Thanks,
Eric

[modok]

I don't know WHAT you measured!
Generally to measure runout the crank would be supported with vee blocks under #3 and #1 main journals. Then you'd read runout on the #2, #4, gear area, and pulley area, and where the flywheel mounts.

The limit should be .002" not .012"

If it is bent it can be straightened.

Recently, I decided to measure the runout on the dowel pins, after I experienced a china crank that had the pattern off center.

[303epps]

Yea I could have descibed how the measurements were done.

I used the same method that Jim Martin described in his engine building article. Used bearings 2 and 4 to check number 1 and bearings 1 and 3 to check number 4. Used a .0005 dial indicator to confirm method and results 3 times.

Called DPR, they suggest grinding as they couldn't guarantee the crank wouldn't break if they tried to straighten it..

[modok]

ah, ok
It can be straightened. easy 15 minute job, not worth shipping cost

how about mounting the flywheel? That's usually a larger mess....

[Brian71]

It's generally a good idea to indicate the journals supported by the Vee blocks first each time you do a setup. If they read anything other than zero, your readings everywhere else on the shaft are compromised.. Which then raises the question, is what you are reading a bent shaft or an out of round journal?

[mark tucker]

and be sure the crank is forward agnist the thrust bearing...and also be sure the thrust bearing raidi is bigenough to clear the bigger raidi on the forged cranks... some are some arnt.if it rides up on the raidi it will show bent but aint. well possiably aint.

[303epps]

Good point on the journal tolerance. Journals measurements varied .0002 -.0003 with mic and did the same with the dial indicator on the stationary bearing.

Mark, I'll recheck bearings 2 and 4 with crank against the thrust, thanks.

Eric

[modok]

IMO DPR does a great job. Good job, good guy.
That said, if anybody tells you they can't straighten a crank that's bent .0015, then, it's total baloney. Of course he can.

It might be a shorthand way of telling you he's not interested in straightening a chinese crank, and the possibility of being blamed for some unrelated problem that could occur later, so instead of just straightening it, he'd rather make sure it's right, which will take more time.

Do you want to learn how to straighten it yourself?

[303epps]

SURE DO!!

[chrisflstf]

BFH and a good size punch. Couple wacks will do it.

[modok]

Before you can do anything, you'll need:
a HARD chisel of good quality steel, then instead of a point, put a smooth raduis on the end. This radius needs to be just tighter than the radius at the corner of your crank journals. Grind to get close and finish with a file. Set it on the crank and eyeball to make sure it's right. this will be used to peen the radius where you want to bend the crank upward toward where your working it. Work the inside half of the raduis, as the outside don't really cause anything to happen.



Will need a 2-3LB hammer. Support the crank in a case half on 1/3 should be ok. Mark where the LOW spots are on the crank and find where the bend IS. Based on your readings it is right in the middle, but which direction?

Stock VW and corvair cranks are probably the easiest to work with. First time I tried a corvair I bent it 3x too much on the first try,,,,,oops. Lot easier to work than larger harder cranks! the 4340 crank will be harder than stock I expect, but still easy as cranks go.

[303epps]

How's will this chisel work?

Eric

[modok]

looks good.

What seems to work well for me:
Once you have determined where the bend is and which spot to work on
Mark where that is.
Start about 45 degrees to one side of the mark and work back to center, then do the same on the other side, so you end in the middle both times.
Maybe... six hits on either side. Light at first, and see if anything happened.
lather rinse repeat, until satisfied. It's great fun, until your half deaf and fingers locked up.

[303epps]

just to confirm - low side down. Correct?

Eric

[modok]

No

Work the concave side.

What if you took a piece of sheetmetal and hammered on it? what will happen? the spot you hit will be flattened and expanded, hammered thinner, or you might say stretched. It will stretch the area. That is what this does is stretch it. If you try to physically bend a shaft you are trying to "stretch" the short side, well, that can be a problem because just stretching a spot can likely make a crack at the surface. BUT, if you PEEN the surface to expand it there is no stretching involved on the surface, rather compression. INSIDE the crankshaft is getting stretched but that's ok, cracks don't start inside.