chunky

I know there's not a whole lot of k20 action in the Atlanta area, especially naturally aspirated, but I assure you, I own the dubious distinction of having blown up more k20 motors than anyone else in the state. Along with that comes rebuilding experience, and I'm going to share a bit of my latest rebuild in this thread.

After 20,000+ miles, my 230ish hp 9000+rpm k20 bit the dust. More specifically, the cylinder #3 piston bit the dust.

What cylinder #3 looked like


What the rest looked like. You may be thinking to yourself "my, what a clean burning engine he has!" Why thank you, I like to think I know a thing or two about tuning.


Rods & pistons laid out


In case you wondered why they're called "H" beam rods.


Long story short, these were CP 12.5: pistons with a thermal barrier coating on the top and anti friction coating on the skirt. I was planning to run a healthy shot of nitrous, but I never got around to it. The pistons were installed with a piston to wall of 0.0035" or 0.0036" which is okay for an n/a motor. However, what happened in the 20+K miles that she lived is that the piston to wall clearances never tightened up to what they should have been when the engine was at operating temperature. As a result, the pistons were loose in the bore and the bottom edge of the skirts all showed irregular wear. Cylinder #3 was just the first to go. The skirt fell apart and then the piston turned sideways in it's bore.

So based on that experience, I decided that this time around I'd go for pistons made of a lower expansion alloy. The only off the shelf pistons for a k-series made of 4032 are the supertech.

Here's some comparison pics of the CP pistons vs. the supertech pistons.

The CP has a much more nicely sculpted dome with a wider peak. However, the Supertech has visibly deeper valve reliefs and yields the same 12.5:1 compression. Just from the way the top of the CP piston is machined versus the Supertech piston, it's not difficult to see why there is a significant price difference between the two brands.


The Supertech has a wider skirt that is also a little longer. The crown of the piston is also a little thicker. The supertech piston also has less taper from the crown to the bottom of the skirt. This makes sense as it is constructed from an alloy that expands less as it heats up.


On the underside of the piston, it's again clear why CP costs more. It's got much nicer machining on the underside of the piston. Dual wrist pin oiling holes are present on the CP piston. The supertech piston does have a thicker skirt. The wrist pin is also shorter on the supertech piston. The CP piston will accommodate a much wider rod, although that doesn't matter much if you are using stock width crank journals.


I got the block back on Friday and measured it out today.

All cleaned up.


The plateau hone looks really good.


My inexpensive bore gauge that gets the job done. I have to do some interpolation to read between 0.0005" increments. Also, the reading shown here is 0.00225" over since the gauge zeroed out on my piston measurement halfway between zero and one mark to the right.


So, the other thing I wanted to do tonight was get a better idea of if the piston to wall clearance I specified to the machine shop was suitable or not. Last night I made up a worksheet in MathCAD to calculate the expansion of the pistons based on coefficients of thermal expansion I looked up for 4032 and 2618 alloys. Those numbers indicated I'd end up with a hot clearance of 0.0011", which seemed reasonable. However, I grew up watching mr. wizzard and bill nye the science guy, so I had to take things a step further and determine the CTEs experimentally.

MMMMMMM, pistons, they're what's for dinner. Note that I marked the measuring points with a marker to ensure measuring precision. Supertech pistons on left, one CP piston and one OEM piston on the right.


They're cooking now!


After crunching the numbers, results were actually reasonably close to what the values I looked up were. Well, for the 4032 and 2618 alloys. Obviously there was no table with specific information on CTEs for the OEM piston material as CTEs for cast material vary widely depending on the specific way in which the piston was cast. But I'm getting sidetracked.

1. CP CTE: 25.7 versus 24
2. Supertech CTE: 23.2 versus 21
3. Stock piston CTE: 16.9

As you can see, the 2618 alloy does in fact have a higher CTE than the 4032 alloy. As expected, the CTE of the stock piston was much lower than with either of the forged pistons. The mathematical model says that the OEM piston to wall in a hot motor would be 0.0002" to 0.0010". With the piston to wall I specified on this motor, my hot clearance should be right in the middle of the OEM range, 0.0006". In short, I chose wisely. Piston to wall is tight enough to help maximize longevity, but not so tight that I'd have to worry about seizing the motor if I run her hard on a hot day.

For kicks, here's a screenshot of my MathCAD worksheet.


I also measured the expansion at the piston crown. The aftermarket pistons generally expanded a good bit more along their thrust faces at the crown than along the wrist pin axis. The CP pistons were within 0.001" along each axis. the Supertech pistons had an average difference 0.006" between the two axes, with the longer axis being the thrust face of the piston. The stock pistons were slightly oval as well having a difference of 0.002" between the two axes, with the longer axis being the thrust face of the piston.

What I find interesting is that although the CP pistons were more round at the crown than the stock pistons, the stock pistons had much more even skirt wear. This highlights two factors that set apart truly great pistons from the rest.

1. having a truly round shape at temperature
2. having a skirt profile that matches the bore at temperature

Now, I'm not sure how the Supertech pistons will stack up, but it would seem that the CP pistons left a little to be desired in the piston skirt profile. Makes you wonder why some companies charge $1600+ for a set of four pistons and what kind of intrinsic value is engineered into the product.

That's it for now. There's a lot of work left to do before I get this motor together. Hopefully, she'll be stronger and longer lasting than the last.

Combustion Contraption

This is one of the most intelligent things I have on HT in a long time. It would stand to reason no one has responded yet and I am the first one to reply, as they are all busy making "wut camshaft/how high a compression ratio can i run" threads. Props to you.

PuroMotor

I wont play and say I understand all your numbers on your worksheet. But ill say that your a man with attention to detail. Awsome research. thumbs up

Nerologic

Everyone is too busy in GDD, they forgot about this forum.

I don't want to complain about "back in the day" everything was better, but back then people looked into these sub-forums and actually found "tech." Now as you can see, these type of threads are very few in between...

Props to the OP

Team Evol Tyrone

Chunky some I used to see in Cylinder three was temp was always hotter so I took my ring gap numbers and went up about 2 on cylinder three. Another way you could do it is add fuel to Cylinder three if you had the right EMS. Good thread though wish more people would do write ups like this. Instead every one does my build threads where everythings sourced out instead of them having true knowledge to build it themselves.

chunky

Thanks guys.

In regards to cylinder #3 running hotter, when I had my injectors flow tested & cleaned, I put the highest flowing on on cylinder #3. It wasn't a huge adjustment though. I also set the ring gap a little looser on cylinder #3, but only 0.001"

Ahh well. It sucks that the last engine blew, but I've found that you can learn a lot from what went wrong with a build, and then apply it to future builds. That's what I'm attempting to do here.

JDM JEFFREY

iTrader: (1)

00MRSi

Great information. Thanks for taking the time to share it with us.

What is your day job?

chunky

Job? what job? I'm a grad student, ME.

00MRSi

Ok. I figured you were an engineer..

BuildOff

In an attempt to decipher what the cause and effect were on your last build, can you show the back sides of the bearings, the con-ron/ cap parting surfaces, and the wristpin bore measurements (top-bottom & side-side). With a slightly generous piston to wall clearance of ~3.6 thou, it's hard to believe that the piston to cylinder gap is the downfall of this build. Seems much more likely that there were small amounts of detonation that killed the piston, and therefore it snowballed into what you saw when you opened the engine.

I may be wrong, but that is a slightly educated guess...

On a much brighter note, it is quite refreshing to see a little knowledge being shown by a youngster!

chunky

I looked at the bearings and there was no sign of detonation. The engine was very well tuned and i had actually refreshed the rod bearings not long before the engine let go. I heard a slight ticking noise one afternoon when I started my car between two tour bus sized SUVs. Although I hadn't really driven the car hard that morning, I thought I might be dealing with a rod bearing that was on it's way out. I pulled the bearings and they looked minty. I put new rod bearings in, but the noise did not go away. I know now that the sound was the start of bad things happening with the pistons.

The installed piston to wall was not excessive by most standards, but my best guess at the moment is that it was too much for a coated piston top. With less heat getting into the piston, the resultant lower expansion would translate into a looser hot clearance. So what I've learned is that the cold clearance doesn't mean a whole lot. You have to consider how things look when they get hot.

I measured the wrist pin bores, and they were not elongated. The con rod bearing/mating surfaces also looked good. I have a very high level of confidence that the engine was not killed by knock as I tuned the motor myself (making seasonal adjustments as necessary) and always ran with the factory knock sensor enabled, which is very sensitive. Aside from which, all the other piston tops/bearings looked perfect. If it was detonation of the scale that would cause this, I would have seen damage in more than one cylinder.

mar778c

iTrader: (1)

3.5 thou on a CP piston is pretty big but thanks for sharing the re-build.

flyrod

Nice tech write up! Was there a lot of piston slap on cold startup? Did the knock sensor pick up any extra noise? I was getting a lot of noise on the knock sensor at certain RPM ranges, which I think must be the pistons. (https://honda-tech.com/zerothread?id=2318492) What were the signs of the piston going out besides the ticking? Did you notice anything wrong before the piston went sideways? Good luck on the rebuild!


Modified by flyrod at 11:34 AM 11/10/2008

bjones

informational thread

4g hatch

#3 cylinder always gets the hottest. anyway im glad to see someone in here share some actual all-motor tech. maybe people like me could learn more

chunky

When I put them in, they said that it was at the loose end of what they recommend. If my memory serves me correctly, they recommended a 0.0033" for an N/A motor which according to my model results in a hot clearance of 0.00051" piston to wall. At 0.0036" clearance, the model shows hot piston to wall of 0.00081". Not excessive by any means. I think the problem is that the thermal barrier coating kept the pistons from fully expanding, resulting in a larger than expected piston to wall when hot.

chunky

Nope, no piston slap at all through most of the motor's life. Looking back, the engine did get a little noisier towards the end, but it wasn't bad until i heard the ticking sound. The ticking sound got progressively worse until it was full blown piston slap after about 140 miles. Before the motor actually let go, I knew it was time to park her when the knock sensor went off. I was trying to limp her home (I was 3 miles from home) when she let go. I should have just parked & called AAA.

Basically, it sounded at first just like a very slight rod knock, and then grew slowly into an audible diesel truck sound.

6SPD_EK

iTrader: (2)

very informative thread

makes me want to look into clearances more closely before putting in some swain coated wiseco's in my build.

what thermal barrier coating was on your CP's?

alterdcreations

do you have a oven big enough for the block?
how much do the cylinder walls expand in relation to the pistons?

mar778c

iTrader: (1)

I think those clearances were too big. I built one or two motors with CP pistons and the clearance is normally .00085 to .0009"/in of piston diameter. In the case of a 86 mm bore, should have about .003" piston to wall clearance.

If your wall clearance is too big the piston will not develop the proper oil wedge on the thrust face.

pogeeboy27

iTrader: (3)

I agree. What happened to all the usual people with all their "theoretical" ideas about what went wrong and what to do next? ME ftw!
To the OP, where did you get the coefficients of thermal expansion? The ones online don't seem to list the coefficients of different grades of alloys.

chunky

Yeah, looking back, I would have done things differently. Now, using my model, I'd do a 0.0033" for an uncoated piston and 0.003" for a coated piston.

Hopefully I've picked a good number for these Supertech pistons.

I used http://www.matweb.com

mar778c

iTrader: (1)

What clearances are you shooting for?


Modified by mar778c at 9:07 PM 11/12/2008

Lubo_25

seeing how cp pistons have thin skirts the tbc coatings dont have much to do with them not expanding enough to provide proper sealing...the piston will always expand to its designated alloy amount...all depends on the machine work done...

changing piston alloys will help slightly but to me toleranced machine work is everything