Block out of square
Thanks to the community for all the excellent responses and suggestions. I'll reply potluck style here.
Too bad you don't know the history of the block. If it's .040 off I would imagine there was an issue with whever was bolted to it last.
Yeah, thought the same. Maybe they had it paired up with an automatic transmission and relied on the torque converter to take up the angular slop? Or they shipped it and made a car that perpetually trashed tranny input bearings? Dunno.
I'm not sure how any of this is your problem. Send it back to the builder. Get another one. No matter if they did it, or didn't catch it, it's not the consumer's issue.
Yup. I think that's my go-to, if I can get them to answer the phone. They insisted it was the bell built wrong, said there was nothing wrong with their stuff. That was before I had dialed the back flange, I thought of that a month or two after the last time I'd talked to them. Big question is whether they'll actually stand behind their stuff. Worst case, I've got a local place, I'll take it there and have them disassemble, start from a brand new block, reassemble. PITA... but, hey, rebuilding a car, I signed up for it. :)
Well being a crate motor I am sure it was line-bored, not having the boring bar perfectly horizonal with the block but instead hone with an slight angle would cause this problem. Removing the main bearing caps and using a micrometer and horseshoe style fixture, take your readings off the top of crank journals, this can tell you for sure. The picture is for reference only...…...
That's a thought, easier to get to that to check than having to tear the entire engine apart.
Maybe I missed this OP.... did you put the dial indicator on the back of the block and check the crank flange for runout? Or re-position the dial indicator 90* on the crank flange and run the test again, as verification that it is not the crank flange that is off of the crank axis? It is just not 100% clear that this has been done.
Yup. I've dialed that back flange about 10 times, re-positioned the dial indicator mag mount on the crank flange each time at a variety of different angles. Each time, it reads out exactly the same. I also have mounted the mag base to the block and dialed the flange as the crank rotates. No runout at all there. So it's not a bent crank or a flange with wobble in it.
On the block line that I worked on, they had cast datums inside the outer cup plugs that our details would use to find the center of the casting, then we would machine datum holes on the oil pan surface... After that the machined datum holes were used for locating the block for all other details... Machining is more accurate than castings as there is more variation in castings that machining..
The first thing that we checked when something was off was the cast datums to the machine datum holes to make sure that they were correct... You should figure out where the machine datum holes are and measure from them to find if/where the problem is...
Interesting. Where do I find those? What do they look like?
If your bell housing indeed does not contact the block except at the holes, then I can't see any reason to not use hardened shims/washers. You might want something to seal it up to keep dirt out.
Wondered about that. There's a lot of open space around the edges of the flange, which seems counterintuitive from a cleanliness perspective... But at the same time, there's a 4x4" hole where the clutch throwout arm goes through, so... It's not like it's airtight to begin with. And the clutch is going to produce particulate as it rubs, so it needs a way to vent that out, right? Does having it sealed... even matter?
Is this a non-stock bell housing? Sounds like it if you talked to the bell mfr.
Yeah. It's made by quick time, and it adapts the Mopar small block to a nonstock TKO tranny.
I understand the concern on the pistons being at different heights. No way to know without somehow verifying. If the crate mfr did not catch the initial problem in the crank line, then I sure would not count on them to make sure things were decked in a way to make it even. You mention a dyno test? WHOSE dyno? The crate mfr? If so, how good is that?
Yeah. Crate engine did the dyno test, we did not observe that test. But at the same time-- when they drop the block into the dyno itself, its not like they're mounting it using the bell flange. So, unless they're
entirely disreputable... I don't yet have any particular reason to suspect that they'd have deliberately faked the dyno results.
At this point, I'd check couple of combustion volumes. Maybe turn the engine on the stand so that the plug holes are straight up on one side set the front cylinder precisely at TDC on overlap, and use a burrette with an appropriate oil to measure the combustion volume. The check at the back cylinder on the same in the same fashion. If the crank was inclined as much as you think, you are going to see a lot of CC's difference, like 5 to 10 cc's depending the head chambers, piston tops, etc.
I like that idea. But If the crank is out of line.. Then... They'd have had to have re-decked the block to be square to the new line of the crank?
If what you say here is actually true, the answer is "easy." You are gonna have to machine the bell mounting surface of the block square, PERIOD. One thing you might do is dial indicate off the FRONT of the crank and see how square the crank is to the timing cover / other machined block surfaces. Maybe the line bore "wasn't so accurate????"
That's an outstanding suggestion. I like that. But... How? I can dial the bell flange, because it's almost circular. However, The timing cover flange is oblong. How would I go about dialing it? If I mount and zero the dial, and then turn the crank, along the way, the dial tip would just sweep right off the flange into open space.
There are numerous problems with this situation: As you can see from the pic I drew, if you mount the trans the way things are right now it will cause the trans center line to be misaligned with the engine center line. This is only true "if" the back of the block is actually "not square" with the crank.
Yeah, that's how I'm thinking it is.
The other problem is that if the threaded holes in the back of the block are "square" to the surface (which they should be) they will also be out of square when you thread in the trans mounting bolts (after you machine the block back face square). If the bolt holes are square to the engine block surface and you machine the surface square to the crank center line the threaded holes will no longer be square to the surface. It's really a complicated problem until you find out where the actual problem lies. If the bolt holes are currently square to the block face and you machine the block square, then mount the trans, the underside of the bolt heads (the ones that go against the back side of the trans flange) will be cocked because the trans flange surfaces are parallel but the engine bolt holes are square to a different surface (the old original out of square surface).
Okay, but... How big of a deal is that, really? I mean, if I have a taper in the back of the block that's 40 thou over roughly a foot, then over the .56" width of the head, I'm looking at 1.6 thou of variation from one side of the bolt to the other? Seems like I can just ignore that amount of error. I mean, do I trust my washer underneath the head of the bolt to be flat to less than 2 thou? Do I trust the cast bottom surface of the head of the bolt to be flat to less than 2 thou of tolerance? Do I trust the paint on the bell on both sides of the hole to be consistent to less than 2 thou of tolerance? Maybe I'm thinking about this the wrong way, but it seems like maybe that particular problem is ingnorable?
