Help Diagnose '96 Frankenmotor, Low Power, w/ Torque Graphs

[aschwerin]

24.1 mpg at the second fill-up. I think that's good considering how much I used 3/4 to full throttle.

 

Overall, the rebuild seems to have worked fine, but it will take a long time to pay for itself.

 

Assuming my build is representative, I would not choose to bother with a Frankenmotor unless it happens to be the only good parts sitting around. Might as well just buy a different car, or a turbo engine, etc. if one solely wants more power. But for a regular car, where engine work is not a hobby, at the end of the day its still a 2.5L engine.

[GeneralDisorder]

Better to just use the 2.5 with 2000 to 2004 flat top pistons, 770 gaskets, and some torque cams. The dual cam heads breathe better. People's fear of the 25D is founded on misunderstanding and lack of experience with fixing them properly.

 

GD

[Subaru Scott]

So, GD, you really DON'T hone the cylinders with a new set of rings? I'm just asking because I've never tried that. And it goes against everything I was ever taught. I'm all for it, if it works! You've really had better success with seating the rings on a smooth cylinder?

[GeneralDisorder]

Yep. We don't touch cylinders unless they are really damaged and have to be oversized. I put forged pistons into EJ255 and EJ257 engines almost weekly and we have NEVER had one burn oil. When I do A "full rebuild" we line hone and deck the blocks. We don't touch the cylinders unless boring them over and then we go straight to 100mm.

 

That goes for other types of engines as well. I have talked to drag racers that would re-ring their big block every 50 to 75 passes and they get faster sealing and higher ultimate compression vs dingle-ball honing.

 

GD

Edited May 7, 2017 by GeneralDisorder

[Subaru Scott]

Mmmm, yeah. Never with dingleballs. They will put deep scratches in the cylinders. A nice rigid hone. I don't get crazy, just break the glaze, then put down a nice, even 40 degree crosshatch.

[GeneralDisorder]

There is no such thing as "glaze". This term suggests some type of deposit on the surface. It's a finely machined surface - with final machining done by the old rings. The new compression ring is cast iron and they seat surprisingly fast. I've heard it said that on the first crank they are wearing in and on the second they are wearing out. Ring seating for us never takes more than 50 to 100 miles. Honestly it's usually done after the 20 minute first run (camshaft break in if applicable) I run them medium hard with 5-10 psi boost if applicable for about 200 then tune them for full power after an oil change. NEVER have them burn oil and compression is always within 5 psi across the board. We just put forged pistons in a 2016 Hyperblue STI with 4,000 miles on the car. It puts down 360 AWHP and doesn't use a drop of oil between changes. 

 

GD

[punx2400]

Hey GD,

Glad you schooled us on that whole honing misconception.  I was ready to take my block halves down to the machine shop for cleaning, inspect the bores, and honing if needed.
I was also going to replace my pistons, but I'm not so sure now.  The reason being 2 of them have scuffing on the skirts, which I believe to be normal, whereas 2 have a pretty gnarly abrasion.
Take a look, tell me what you think.  It sounds like I was about to go throw a bunch of money out the window...  

Normal wear I assume
https://ibb.co/cgKFOF

 

Abrasion(can barely feel it with your fingernail

https://https://ibb.co/iq9tAv

 

Other scratched piston skirt, after a minute of Scotch Brite'ing

https://ibb.co/nc00qv

[GeneralDisorder]

Those look fine. Get them knurled if you like - it will help with piston slap. Otherwise just put them back in. I've seen 1000x worse - we just knurl them, and file to fit.

 

GD

[punx2400]

Good.  That'll save me some cheddar.  I forgot to mention the skirt knurling in my last question.  Who can do that for me and what's the ballpark figure on the price tag?

[GeneralDisorder]

Depend on where you are. Have to call some machine shops but most don't have the equipment anymore. We charge $100 for a set of pistons. It is the installers responsibility to file them for proper fit.

 

GD

[punx2400]

Machine shops.  Got it.  Well, I'm the installer, so unless I learn how to properly file to fit, I guess that's out. 

 

Meanwhile, since I'll be going with the original pistons in untouched bores, I WILL need the .25 over sized rings because of wear, not the STD size, right?  The motor has 130k on it.

 

Thanks in advance

[GeneralDisorder]

No. Standard sized rings. Oversized rings are for oversized pistons.

 

Find NPR brand rings if you can.

 

Filing them to fit isn't hard. It's trial and error. You just take off the high spots and then test fit. A little twist and shove will show the high spots with shiny. You want them pretty tight. I keep them slightly interference. They loosen over a few thousand miles.

 

GD

[punx2400]

NPR. I'll do what I can.  If not, factory rings are the next best thing, yeah?

 

Your opinion on ACL Race Bearings? That's what was listed as Premium/Performance on Rockauto.  I'm not trying to build a 6 second car, I just think for the few extra bucks, I could get the good stuff, or what I'm under the impression is good...  How bout con rod bearings?

As far as the skirt knurling, do you have any links to say a tutorial video or a write up or something?  I'm having a hard time picturing just WTF I'd be doing, and can't find anything on it besides pictures.  That, and one's "pretty tight" could easily be mistaken for another's sloppy as hell, or even waaaay too tight.

[GeneralDisorder]

Factory rings are very expensive. They are fine but aren't any better than NPR.

 

ACL race are good bearings but you do want to consider the intended use of the engine. The trimetal bearings (race) are more resistant to abrasion and heavy loading but are less capable of particulate enbedment.... The standard ACL bearings are better suited to non turbo engines in my opinion.

 

You will need to have the block mainline honed, the crank should be turned undersized and mirror polished. The assembly process is not as straightforward as it seems. The clearances are extremely tight and you can't trust the case half dowel pins....

 

Skirt knurling is basically from the 1960's. No one really does it anymore. But in the specific case of the 25D/251 style pistons it really helps with piston slap. I had to buy my knurling machine from a machinery dealer out of state and have it shipped. No one makes these machines anymore. I don't know of any tutorials, etc. I can post pictures of before and after. It's not difficult - you just hand file till you can install them by hand. I doubt you could install them too tight. I put them in VERY tight. So tight that you lose a few MPG till they seat in. If you can get it assembled without spinning it with a 2 foot breaker bar - it's good to go.

 

The bottom end requires a lot more finesse though. That's a horse of a different color. I typically tell people to expect at least a couple failures (rod or thrust bearing) before you get the process perfected. It's not at all forgiving. The clearances are insanely tight due to thermal expansion of dissimilar metals.

 

GD

[punx2400]

what about the bottom end?  Like torquing the halves improperly/using too much fujibond/installing wrong bearings?  You're scaring me here with this "expect a couple failures" business.

[GeneralDisorder]

You need to have an accurate bore gauge. The main clearances are between .0004" and .0012". I shoot for .001" on all my builds.

 

The cases, when they aren't brand new, have VERY POOR assembly accuracy repeatability. Since the mains have a clearance of .001" (or less) when cold, that means .0005" on either side. This means an assembly variation of .0005" will cause TOTAL DESTRUCTION of your engine. Think about that for a second. An average human hair is .004" - a variation of 1/8th the thickness of a single hair will cause utter devastation inside your engine.

 

The steel dowel pins are not accurate enough on their own. Also the repeated disassembly, reassembly process, as well as a hundred thousand+ miles means they can't be trusted. Aluminium has the same density as oak for machinist purposes.

 

You are attempting to assemble an engine made of oak, with the tolerance of 1/8th the thickness of a human hair.

 

Now ask yourself - what, in your life experience, has properly trained you to work in this medium and at these measurements?

 

The rods do this too. You have to measure every rod clearance on three angles not just in the thrust direction. The factory rods have poor assembly repeatability also. You have to see how they want to assemble and then beat on them in the direction that corrects their geometry as you tighten them down in stages.....

 

I caution people about this all the time. There are VERY FEW people in my industry that can do it at all - let alone every time with 100% success. Over the course of my first half dozen engines I quickly adjusted and corrected my process. I lost three engines out of the first six or so. Two to rod bearing failures, and one to a thrust bearing. We have an extremely rigid set of rules and processes that we go through to insure correct assembly and alignment.

 

Once you get past the block half assembly the rest is easy. Everything else is pretty forgiving. The bottom end on a Subaru engine is probably the least forgiving rotating assembly on the planet from a machining and reassembly perspective.

 

Beware.

 

GD

Edited June 15, 2017 by GeneralDisorder

[punx2400]

I like shootin craps.  Eff it.  Roll em.

[punx2400]

Those analogies are pretty kickass too, btw.  

[punx2400]

So I should just sell/scrap my stripped and split 25 and go buy another core, but this time, don't do anything but a ring job, then slap those 22 heads on and go.  Roger.

[GeneralDisorder]

It depends on if you have an bore gauge, micrometers, and are willing to learn to use them for this job. Also need to find a machinist that can do a mainline hone.

 

The process of insuring proper alignment of the case halves during assembly involves noting how the case "wants" to go together and then rather violently "aligning" the case halves using the crankshaft, a dead blow, and a modified torque sequence as you do the final assembly. Basically "walking" the case into alignment using the crankshaft journals. Same goes for the rod bearings.

 

It is also helpful to "read" the wear patterns on the old bearings. This will shows what journals need work and if the crank and rods are straight, or need to be corrected or replaced.

 

GD

[punx2400]

Would you care to take a minute to explain the process, and what these steps and rules are? I'm on the fence about whether I'm gonna roll the dice and just slap the halves together, or take it to a machine shop/builder and have them assemble the shortblock...