The OEM rod bearings debate

[Calum]

Originally Posted by OrbitalEllipses

When the WRX hits the ground we'll find out, we haven't had anyone dissect the JDM engine for us at this point so the WRX is probably when American tooners will try.

Good point, thanks.

[nelsmar]

This whole scenario just makes me feel like its the worst time to build a motor haha. Oh the joys of being an early adopter!

[DAMotorsports]

Having opened a FA20 i would not fear it. i would take your time and go over everything or use a motor builder. setting the clearances not matter which bearing takes time. and for the oiling they are a few thing you can due to improve the psi. knowing the clearances of the stock motor i would not use anything thinker then 0w-20w

[Captain Insano]

Appreciate your posts on this forum as you are really pushing the limits and helping everybody with your info. So don't take this post wrong please...

In my experience (I've had motors go on other cars with rod bearing problems) when a rod bearing "fails" it is typically due to detonation or oil problems (heat or starvation or both) and not solely the bearing. You presume oil issue due to not having detonation, how do you know you did not have detonation? Are you running an engine management system that literally logs constantly? Do you have knock sensors on the car? Sometimes detonation is very hard to hear with human ear when you are running hard due to other noises happening especially on modified cars. Not saying your issue wasn't oil caused, but wondering how you know it wasn't detonation.

Please bear in mind it is way, way easier for a tuner/vendor to presume a problem was caused due to an "oil issue" than due to a "tuning issue". One places blame on them, the other does not.

Not saying this is the case again - but I've had vendors pull the "oil issue" card with me, but ended up seeing detonation evidence on tear down. Fortunately I had an independent motor builder do tear down and I inspected the parts and at that point I could cut my losses and move on. Unfortunately, if the same vendor that tunes your motor tears your motor down after destruction and you don't inspect the tear down they can tell you whatever they want.

Just my wonderful experience with aftermarket tuners. I would say even in the very best situation if you or an independent motor builder tears down your destroyed engine one cannot always tell EXACTLY what the culprit was between all the different things that can be working in conjunction with each other to cause an engine to ultimately fail.

Originally Posted by cf6mech

I do think there is an issue with the OEM bearings at high hp FI applications,...I feel others aren't putting ACL or aftermarket bearing in because they are not available so they roll the dice. I do believe its an oiling issue, which an aftermarket performance bearing addresses with more oiling clearance. The stock bearing is hard, extremely hard, to hard for a performance application in my opinion. I had bearing failure on E85 with no detonation with a very sound tune on a built motor running close to 400whp Using Amsoil 10W 40. I'm now equipped with Accelerated Performance Pauter rods with ACL bearings and at 300whp @ 10 psi with no issues.

[arghx7]

Originally Posted by cf6mech

I had bearing failure on E85 with no detonation with a very sound tune on a built motor running close to 400whp Using Amsoil 10W 40. I'm now equipped with Accelerated Performance Pauter rods with ACL bearings and at 300whp @ 10 psi with no issues.

One thing people don't realize is that when you run E85 or race fuel cylinder pressures can go way up with enough spark advance. I know that sounds strange--say you have 300whp (or 300wtq) on E85, versus similar output on pump. The E85 is much less likely to knock, so it should be better for the bearings, right? Well, sort of.

Peak cylinder pressures is highly sensitive to spark advance, or rather the phasing of peak combustion pressure. Allow me to illustrate.



Here you can see a 300 engine cycle average pressure trace on this particular direct injected engine at an idle condition where knock is not really going to happen. This is all at constant torque, so other things were adjusted to make up for the effect of changing spark advance.

The top pressure trace has peak pressure occurring around 40 degrees ATDC firing. Spark is very late, after TDC. Now look at the Y axis. Peak pressure is about 9 bar.

Now look at the middle trace. Peak pressure is around 15-20 degrees ATDC, which at this speed and load is roughly MBT (minimum spark advance for best torque). Peak pressure is at 12.5 bar. Do the math--that's about 50% more cylinder pressure from retarded spark vs MBT, but engine output in the case is the same.

Finally, check out the bottom trace. This spark is way past MBT, something like 30 degrees BTDC. Peak pressure is about 15 degrees BTDC, way too early. Peak pressure is now about 19 bar. From the two extremes of unnecessarily retarded combustion to overly advanced combustion we double the cylinder pressure. Let's look at it with a pressure-volume diagram, (x axis volume, y axis pressure)



This relationship between non-knocking spark advance and peak cylinder pressure also exists at high loads. So what does that tell us? When we make our engine output relying on using knock resistant fuel and spark advance, it greatly increases cylinder pressure. What do you think that does to bearings? I'll say that again: Fully taking advantage of knock-resistant fueling with spark advance can be very stressful on the engine due to high cylinder pressures, and it's not knock causing the stress.

This engine was never designed for high loads (lots of airflow crammed in there at a given rpm) and high cylinder pressures. Think about it when you put E85 in there, boost it, and have a low knocking tendency fueling (race fuel, E85, maybe water/meth).

[Captain Insano]

Thanks for this post. Sorta reinforces the last paragraph in my post before. Typically detonation or oil for bearings, but at the end of the day there are so many things it could be especially when considering combinations of things and all those possibilities. In this case you are saying possibly just the nature of much higher cylinder pressure, and I agree, that would stress the bearings more.

[nelsmar]

@arghx7 Thanks for the detailed post. I have always wondered how much of a pressure difference there was on spark timing. Do you have a reference for those photos? I would love to read more.

[mike the snake]

So when tuning, do tuners tune for power by adding timing and simply listening for knock?

I'm wondering if or how they know to tune for the best, safest power.

[arghx7]

Originally Posted by nelsmar

@arghx7 Thanks for the detailed post. I have always wondered how much of a pressure difference there was on spark timing. Do you have a reference for those photos? I would love to read more.

Here's something to check out. It's from a Ford study of using dual fuel pump gas + E85 capability in an 3.5L Ecoboost engine with PFI and DI. Ford Ecoboost E85 Study

Here are some more relevant charts.



For purposes of this discussion, BMEP can be converted to torque. 10 bar BMEP is 260 lb ft and 20 bar is 520 lb ft. CA50 is location of 50% burn. MBT is typically 6-10 degrees after top dead center. Add roughly 5 degrees for location of peak pressure (CA50 of 10 = peak pressure at roughly 15 degrees ATDC firing).



Example: near 20 bar BMEP on this engine, so 520 lb/ft of torque at a fixed 2500rpm. With pump gas only, sprayed through port injectors, we run -10 BTDC degrees spark (10 degrees ATDC). With E85 direct injection we can run 30 degrees spark.

The second chart is kind of complicated and hard to explain without reading the attached paper. Basically, they were dialing in spark, AFR, and E85 blend % to #1 not overheat the turbo and #2 not exceed 100bar peak pressure in order to protect parts in the engine. But you can see that they limited peak pressure to 100bar even when the engine wasn't knocking in order to prevent damage. They pulled timing even when they technically didn't have to, which then fed into the E85 blend ratio and Lambda so that the turbo didn't overheat. It's also a little complicated because you can go by average peak pressure across cylinders, average peak pressure over some number of cycles on a single cylinder, or peak pressure on a single cycle for a single cylinder.

[Guff]

Sub'd

[nelsmar]

Very interesting. Where can I find the first article? I am quite interested in the way they handled the torque compensation. I also would love to see the comparison of ignition vs AFR to pressure.

If I'm understanding correctly the reason the 500 torque is still driven is due to the "usable" cylinder pressure. As if you were to create massive pressure prior to TDC the duration of the high pressure is wasted during compression, and only partially usable post top dead center. This would visually show the same power output due to massive cylinder pressure and "shock" due to the "spike" of which the cylinder pressure increases. So you may be putting literally 2x the stress & at 2x the shock speed that would translate to possible bearing shift, or other issues due to the requirement for the components to respond to the pressure.

We constantly just think about power output and the spark as being "instantaneous" but the pressure has a duration. You could make 20bar pressure peak at BTDC, but by the time the stroke hits TDC you are only seeing 10Bar, both of these instances would see the same output. But the prior scenario would be showing 2x the stress internally.

[nelsmar]

I would be really curious to see what kind of ignition advance @cf6mech was having when his forged engine was struggling. If the engine was tuned past MBT the cylinder pressure could be significantly higher... However I don't want this to turn into a debate about who is a good tuner, and why. But he is one of the only to speak up and share his experience at this point. I don't think his crank was ever inspected he just replaced it but i cant verify that.

Would having a softer bearing be more resistant to damage via aggressive cylinder pressures by having more "flex" ability in them? If so that could explain the issues being seen. I have thrown a rod on E85, but have never taken the time to inspect it. I am curious to see what my bearings & crank look like. In the next few weeks ill inspect them. I also have logs of 'similar' scenarios showing my ignition timing and other variables that may be of use.

I really wish I had a setup locally that I could test on a dyno to see exactly where MBT was compared to my vehicles tune. I really want to find out what the limits on the OEM bearings really are as Don @ accelerated's solution seems to be the only available after market option at this time. And we have a number of people in the 400-600HP club on OEM blocks running just fine at this time. Although I am sure it is on borrowed time and for all we know it could be due to production inconsistencies. Does anyone know anyone that had an engine fail running on pump gas where you would actually see detonation? Everyone that I personally know that has thrown a rod or had a drastic failure was on E85. In which you may not see knock... you may just see a bent rod due to cylinder pressure.

Again the reason for this debate is a number of shops are saying the OEM bearings are just fine (such as crawford with over 15k miles on OEM bearings with forged internals). As well as a number of other builders who say they prefer OEM bearings when possible. Just so few people with built engines have popped their head up and shared their experiences.

Personally i'm starting to lean towards doing my build with OEM bearings at this point and tuning extremely conservative on ignition timing during any load beyond highway engine loads.

[arghx7]

Originally Posted by mike the snake

So when tuning, do tuners tune for power by adding timing and simply listening for knock?

Adding timing, or adding boost in the case of forced induction. You can listen for it but most of the time with a stock ECU it's just a matter at looking at what the knock control system is doing

I'm wondering if or how they know to tune for the best, safest power.

So there are two things we're talking about here. One is knock, which will cause pressure spikes, increased temperatures, etc. That is definitely going to cause problems for core engine components like pistons and maybe bearings in this case.

The other thing is essentially pushing the engine too hard without knock. At the risk of oversimplifying things, you can make more output by throwing more air at it and by adding spark if you aren't knock limited. Both have advantages and disadvantages but they interact.

Judging whether the tune is safe even in the case of no knock is not easy. No matter what there is trial and error involved. It's easier to understand if you have some specification for the parts. We can certainly guess that the engine can handle a max speed higher than the factory rev limiter, and there haven't been widespread concerns about tunes that have revved higher.

Load on other core components from pressure and temperature are difficult to judge though. You need some specification on what they can handle, which you'll never get for the OEM parts but you could get for the aftermarket maybe. You also need very expensive instrumentation (parts and labor involved) to read temperatures and pressures in the combustion chamber, or in the oil galleries.

So really all you can do is trial and error for this stuff. All the older platforms have gone through it already. Look at some forum posts for DSM's or EG/EK Civics from 10 years ago. People were working through the same type of issues.

[Dezoris]

Well lets face it too, most engine builders dont really share out or volunteer information. Its bad enough trying to get vendors to say anything anymore about testing. But one thing is for damn sure those claiming 400ish HP on here you wont see any logs or much proof of any sustained abuse like track time.

I mean you can throw boost and crazy timing at any motor for short periods but sooner or later something is going to give. I dont take any of these shops claiming high HP seriously. Namely if its just dyno speak. Nelsmar keep doing research and post back.