Pedros Techno DOF IMS fix?

[Jager]

Originally Posted by Jake Raby

The guys in my class today, here in Atlanta will learn about this tomorrow.

Otherwise its not shared or even spoken about. Plus, I want to see how long it takes the "expert opposition" that has attempted to join our ranks in the past year or so to finally figure it out. We've only known about it for a decade.

I am currently attempting to see if this was part of the discovery documents for the IMS Bearing Failure class action lawsuit. I'd be willing to bet that its not.

Cracks in the bearing cage? Various frequencies created as a result?

[Jake Raby]

Originally Posted by Jager

Cracks in the bearing cage? Various frequencies created as a result?

Nope.. But this could increase load on the cage and lead to premature failure :-)

[Jager]

Originally Posted by Jake Raby

Nope.. But this could increase load on the cage and lead to premature failure :-)

Are you seeing a build-up of some kind on the bearings??

[moresquirt]

Jake,do u believe that regardless of how well the bearing is lubricated (ie DOF) the inevitable is bearing failure eventually regardless off single ,double ect. From what ive read
it would appear that lack off regular oil changes,wrong oil viscosity,driving at slow paces in higher gears,staying at 3000 rpm for extended intervals ect all will contribute to earlier failure but basically at standard bearing in this application is a bad idea and nothing but taking the bearing out off the equasion is the only SOLUTION!!

[Jake Raby]

Originally Posted by moresquirt

Jake,do u believe that regardless of how well the bearing is lubricated (ie DOF) the inevitable is bearing failure eventually regardless off single ,double ect. From what ive read
it would appear that lack off regular oil changes,wrong oil viscosity,driving at slow paces in higher gears,staying at 3000 rpm for extended intervals ect all will contribute to earlier failure but basically at standard bearing in this application is a bad idea and nothing but taking the bearing out off the equasion is the only SOLUTION!!

Based on my direct experience the problem is the application of a ball bearing. The constant that I have mentioned in this thread applies to every IMS Bearing equipped engine from 1997-2008.

Opinions will vary and mine is very biased. Some contend the exact opposite of this.

The attendees of my class just learned of this constant today. Their jaws dropped when I held the bearing in my hand and illustrated just one thing. I could see the "dots connect" pretty quickly, then came the comments.

Tony is going over the IMSR procedure now in class :-)

[Jamesp]

Jake,
Any idea of the typical run out on the IMS driven gear? I had the failed IMS from my engine measured at 0.015. Quite a bit for a gear, but as it is a chain drive if may be ok. In any event that was a major driver in replacing the damaged shaft with an undamaged one. My thought at the time was if the IMS gear run out was poorly controlled in the design, additional side loads could shorten bearing life in high run out cases even with a deep groove ball bearing.

[Jake Raby]

Originally Posted by Jamesp

Jake,
Any idea of the typical run out on the IMS driven gear? I had the failed IMS from my engine measured at 0.015. Quite a bit for a gear, but as it is a chain drive if may be ok. In any event that was a major driver in replacing the damaged shaft with an undamaged one. My thought at the time was if the IMS gear run out was poorly controlled in the design, additional side loads could shorten bearing life in high run out cases even with a deep groove ball bearing.

Good thought that few consider. The acceptable runout is less than 1 thou. for my engines.

Too much runout, and the associated timing chains see accelerated wear as well as the replacement bearing. This is one reason why we highly discourage cutting the race of a failed bearing out of a shaft and resurrecting it, because the heat and trauma associated with bearing failure often creates runout that can lead to shaft separation or future bearing failure.

One supplier of "rebuilt shafts" is selling shafts with another bearing design that have previously failed and have been "repaired". We'll see how that works out for them and the buyers.

[southernstar]

Jake, without getting into semantics, surely you must agree that the problem is a combination of the original bearing design/s AND inadequate lubrication. Indeed, even on your own site you highlight the same under the heading IMS Failures Explained:

"Next, while the grease seals may still be intact, the permanent lubricant inside the housing has been dissolved. The compromised grease seal now provides a barrier, preventing the seeped-in engine oil from leaving the chamber, but also preventing enough fresh engine oil to enter that might cool and lubricate the bearing surfaces." Then "as inadequate lubrication, extreme heat and corrosion continure, mechanical deterioration accelerates."

IMS Failures Explained

I can only assume that you continue to believe, based upon your likely unparalleled experience in rebuilding M96 and M97 engines, that these are amongst the stages of IMS failure and that inadequate lubrication is a significant contributing factor.

What can we safely conclude about IMS ball bearings?

1. BALL BEARING IMS BEARINGS CAN WORK:
We know this because the vast majority of M96 and M97 engines do not suffer IMS failure. Many have gone well over 200,000 miles on their original bearings (both double-row and single-row) without suffering a failure. We must keep in mind the numbers released by Porsche and the fact that they are based upon hundreds of thousands of cars produced and up to 17 years of real-world use.

2. SOME SEALED FACTORY IMS BEARING DESIGNS ARE MUCH MORE RELIABLE/DURABLE THAN OTHERS:
- based upon the data obtained from Porsche during the discovery process in the class-action suit, the original dual-row bearings failed at a rate of much less than 1% regardless of mileage, use and maintenance schedules, whereas 8 to 10% of the original single row beaings failed.

3. CERAMIC BALL BEARINGS ARE MORE RELIABLE/DURABLE THAN STEEL BALL BEARINGS OF THE SAME DESIGN:
- As you indicate in reference to your own ceramic IMS bearing retrofit (and as confirmed by testing and years of experience in various applications), properly engineered ceramic ball bearings are more durable/reliable than properly engineered steel ones.

4. ALL BALL BEARINGS NEED ADEQUATE LUBRICATION:
Trite, of cousre, but important. Porsche erroneously chose not to lubricate their IMS ball bearings with oil, but relied upon a seal to keep the bearing packed in grease. As you highlight on your own site, as the seals start to fail on these bearings, sufficient oil is allowed in to wash out the grease, but insufficient to allow adequate lubrication of the bearing by oil. Ultimately, the result is failure of the bearing. Pedro echoes your observations in this thread when he says: "The oil that seeps into the IMS is not replenished and that's why (the bearing) is cooked."

5. THERE ARE TWO WAYS OF IMPROVING THE LUBRICATION TO THE BEARING:

The most simple of these is by removing the outer seal to the original IMS bearing - something which you and many others recommend as a stop-gap measure for those with the later, larger single-row bearing who cannot afford an engine tear-down in order to upgrade the bearing. Indeed, it is also why your (and Pedro's) ceramic retrofit bearings are unsealed. This method eliminates the grease and relies upon splash oil to lubricate the bearing. A question remains, however: is splash oil lubrication adequate in the long run, regardless of how the engine is used? There is insufficient data at this stage to answer that question definitively and, indeed, you recommend replacing your own ceramic bearings every 50,000 miles or 4 years.

The second method of improving the lubrication to the bearing is by DOF (direct oil feed), rather than splash lubrication. This is the method extolled by Pedro. Clearly it will provide much more oil and lubrication to the bearing under all driving/idling conditions than splash lubrication. Using it will undoubtedly eliminate the inadquate lubrication problem that you and others highlight as one of the stages of IMS failure.

Will DOF cause inadequate lubrication elsewhere in the engine? As we all know, oil pressure varies wtih oil viscosity, oil temperature, rpm's and engine wear. The M96 and M97 engines were designed and built to take this variation into account and there is no reason to believe that directing a small amount of oil to this bearing will reduce oil pressure elsewhere to a point of inadequate lubrication ( unless, of course, the engine is already suffering from low oil pressure). Indeed, you include direct oil lubrication to the bearing in your IMS Solution and must have come to the same conclusion.

Will DOF cause increased aeration and foaming in the oil that modern oils are incapable of coping with? Comparing the oil on the dipstick of a DOF equipped car versus a non-DOF equipped car after a hard run should go a long way towards answering this question. I, for one, encourage Pedro to produce a video checking the oil of two comparable cars (same vintage, mileage and with fresh oil of the same type) after comparable on-track drives. Common sense, IMO, also suggests that, after the initial start-up of a DOS system, air cannot enter the bearing housing as it is a sealed system: by definitioin, without air there cannot be aeration of the oil.

Is the IMS solution an upgrade over the original sealed IMS bearings? Undoubtedly. Is it also a significant upgrade over a properly lubricated, unsealed, ceramic ball bearing IMS bearing? Possibly - although in the case of dual-row bearing engines with upgraded ceramic bearings and improved lubrication, the 'solution' will have to be unbelievably durable to get better results over time.

Considering the cost, at this stage I still believe that using an LN ceramic bearing and splash lubrication is the way to go for most people - certainly for those with dual-row bearings. I am also inclined to believe that for the larger single-row bearing equipped cars (where the bearing cannot be upgraded without engine tear-down), removing the outer seal is an essential first step. However, for these engines, I believe that DOF is the way to go. Whatever theoretical risk there is of aeration should be overcome by dramatically reducing the known risk of IMS failure due to inadequate lubrication. Finally, with more evidence on the issue of aeration, I would also think that DOF and an upgraded bearing would be the most cost-effective fix for the remaining cars, rather than merely an upgraded bearing.

If money is no (or virtually no) object, then I would go with the LN Solution. History from Porsche's own non-ball bearing IMS cars has proven it to be a reliable component design that should all but eliminate the worry of IMS failure. If only the M96/M97 engines had been designed that way in the first place!

Brad






-

[Jake Raby]

IMS bearing failures have dozens of contributing factors that we have discovered. There's probably just as many that we haven't discovered yet.

If we knew all about it, what I do wouldn't be called research.

[Kirk]

Southernstar... wow great summary and analysis!

I can buy the multiple failure mode theory. Seal failure leads to a lack of lubrication. A lack of lubrication leads to increased heat, pitting, wear, failure. How much surface area you have affects how quickly you realize failure - single row bearing thus will fail quicker than a double row, etc. The argument back and forth about whether it is a bearing load failure or a lubrication failure sounds a lot like the argument about which came first, the chicken or the egg. I looked at LN's website and see that they recognize the lubrication issue certainly as a contributor. It does seem then like the best solution would be the biggest, strongest bearing that you can fit with good (DOF) lubrication.

Personally, I understand the bearing weakness argument, but I just don't buy it. It seems to me that a steel bearing with proper lubrication should last just fine. That's my opinion and I think that's what I'm going to do with my car - steel bearing, outside seal removed, DOF. Will I still change the bearing with clutch changes? Honestly, yes I probably will, but mostly because it's just cheap insurance for me personally. I do my own work, so the extra labor cost is zero and the time won't be bad. I also buy a LOT of bearings at my work and can get the steel bearing for the same cost as a good lunch.

I'm not really concerned about oil getting into the shaft. Most people removing their IMSB have found oil in the shaft. It happens. It doesn't seem to be an issue. With the DOF at least that oil would not be stagnant.

I'm not really concerned about aeration either. If you look at the videos and good pictures of where the IMSB is installed you'll understand that it is a very, very tight place. That's why I don't like the splash lubrication approach... I don't really see a good path for oil to get into the bearing. Sure, maybe while the car is sitting and oil has drained back and filled the sump. But with the car running, oil pumping sloshing and moving about, I don't see a consistent, reliable way for oil to lubricate the IMSB. That's why LN uses a high wear bearing. The beauty of DOF is that it takes a lot of variables in lubrication out by providing a consistent flow. But you're pumping this oil into a very tight space. What's going to happen? The void is going to fill with oil, you'll probably even get some backpressure in the oil hose, and the excess will seep out of the small gap between the shaft and case. How is a lot of air even going to get in there? The bearing will be flooded and it should be happy.... then I'm happy.

Kirk

[Kirk]

Originally Posted by Jake Raby

That single factor is what makes the difference. Every single Bearing Engineer we have consulted with has brought the point up, but we considered it before they did.

So Jake... if we guess this super secret factor will you confirm if it is correct or not? I have seen Charles Navarro of LN state that he would take this secret to his grave!

I am going to throw out my guess and I am going to bet that I am spot on.

The issue is that in the design of the intermediate shaft the ball bearing application is compromised (not optimal) and the magic number (or constant) that it is compromised by is 20%. We state load numbers for the 6204 bearing based on manufacturer's numbers, but those numbers assume that the inner race of the bearing is turning. In most applications the inner race is fitted to a shaft through an interference fit, the shaft rotates, the inner race rotates, and the outer race is held in place in a fixed housing. Bearing load numbers are based on this scenario.

This is not what happens on the intermediate shaft though. That scenario is exactly the opposite. The inner race is fixed and it is the outer race that is pressed INSIDE the intermediate shaft that rotates.

So who cares if it's the outer race that rotates rather than the inner race? Engineers care because this one little difference has a significant impact on how much load the bearing can take. The general rule that is used in bearing calculations is that the impact is a 20% decrease in load. Decreased load effects calculated bearing life and failure rates.

Is this significant then? 20%? Damn straight that is significant. So how to address this? LN went to a stronger ceramic bearing with a much higher load rating than OEM. But eventually LN went to a plain bearing with significantly different load characteristics altogether.

So what do you say Jake? Do I get the gold star??? :dance:

Kirk Bristol

[Jake Raby]

But eventually LN went to a plain bearing with significantly different load characteristics altogether.

LN Engineering didn't...

IMS Solution LLC did. LN doesn't sell the IMS Solution, neither does Flat 6 Innovations.

[Jamesp]

Kirk,

Thanks for pointing out the bearing life factor. Do you have independent calcs on IMS bearing life utilizing the 1.2 factor? I'm deciding what to do with my IMSB right now and have not had time to do more than look up the 1.2 factor for rotational arrangement. I'll run the calcs later, but would like to see a second set to verify my work. This may have a big impact bearing selection.

[Kirk]

Originally Posted by Jake Raby

LN Engineering didn't...
IMS Solution LLC did. LN doesn't sell the IMS Solution, neither does Flat 6 Innovations.

So you are saying that LN and Flat 6 had nothing to do with the development of the IMS Solution??? Just because you started a LLC to cover your other businesses from liability lawsuits means very little if it is still the same people (George Navarro and Jake Raby).

Jake, you didn't answer my question though, do I get the gold star for exposing your super secret, the one that Charles was going to take to his grave???

I can see how this is significant. It should have had an impact on the class action suit against Porsche as their DESIGN is flawed and compromises the bearing! I can also see why Feelyx's idea got bought out for a patent. Yes, he was working on a direct oil feed system, but that's pretty straight forward. The main project he was working on was moving the bearing out of the shaft. He was focused on using a bigger bearing, but his design also moved the rotational point from the outer race to the inner race. I don't know if he realized that this one change would increase the life of the bearing by 20%.

Who has done an IMS change (New Oil Fed Design Idea) - Pelican Parts Technical BBS

Kirk Bristol

[cnavarro]

Originally Posted by Kirk

So Jake... if we guess this super secret factor will you confirm if it is correct or not? I have seen Charles Navarro of LN state that he would take this secret to his grave!

I am going to throw out my guess and I am going to bet that I am spot on.

The issue is that in the design of the intermediate shaft the ball bearing application is compromised (not optimal) and the magic number (or constant) that it is compromised by is 20%. We state load numbers for the 6204 bearing based on manufacturer's numbers, but those numbers assume that the inner race of the bearing is turning. In most applications the inner race is fitted to a shaft through an interference fit, the shaft rotates, the inner race rotates, and the outer race is held in place in a fixed housing. Bearing load numbers are based on this scenario.

This is not what happens on the intermediate shaft though. That scenario is exactly the opposite. The inner race is fixed and it is the outer race that is pressed INSIDE the intermediate shaft that rotates.

So who cares if it's the outer race that rotates rather than the inner race? Engineers care because this one little difference has a significant impact on how much load the bearing can take. The general rule that is used in bearing calculations is that the impact is a 20% decrease in load. Decreased load effects calculated bearing life and failure rates.

Is this significant then? 20%? Damn straight that is significant. So how to address this? LN went to a stronger ceramic bearing with a much higher load rating than OEM. But eventually LN went to a plain bearing with significantly different load characteristics altogether.

So what do you say Jake? Do I get the gold star??? :dance:

Kirk Bristol

This is just another piece of the very complicated puzzle. When we first started looking at this problem, the use of the bearing in outer race rotation was one of the first problems I knew exactly how to tackle. That's the reason for a ceramic hybrid bearing with sintered silicon nitride balls to be specific.

Jake and I have been using sintered silicon nitride for various components including lifters in pushrod Porsche engines primarily for their wear performance, but the added weight savings is huge.

Now consider the reduced mass of the ceramic balls and how that relates to the loads exerted on the races during the constantly varying acceleration and deceleration of the IMS bearing. This is even more important considering the load on the IMS bearing isn't even across the whole circumference of the races. Since the sintered silicon nitride balls are only 40% of the weight of equivalent steel balls, centrifugal force is lower - the lower weight of silicon nitride balls enables rapid accelerations and decelerations with reduced wear.

At any given time a very small surface area of the bearing is carrying all the load. To simplify what's happening, the balls are slung round 3/4 of the circumference of the bearing and they go from unloaded to fully loaded with only 1 to 2 balls carrying all the load. This is why we see flat spotting of the balls and skidding in the races leading to pitting and eventually complete bearing failure.

Integrated Reliability Solutions

Considering the bearing is in outer race rotation and already prone to skidding, this is just one of the reasons we chose against roller bearings from the very beginning and focused on ceramic hybrid bearings.

However, I will clarify that just going to a ceramic hybrid doesn't give you a higher load capacity than an equivalent conventional bearing but their benefits certainly outweigh the cost.

Hybrid Ceramic Bearings | Applied.com

PS. George is my dad. He used to work in the aerospace field in sales and he came to LN a few years ago to help my wife and I.

[Kirk]

Originally Posted by Jamesp

Kirk,

Thanks for pointing out the bearing life factor. Do you have independent calcs on IMS bearing life utilizing the 1.2 factor? I'm deciding what to do with my IMSB right now and have not had time to do more than look up the 1.2 factor for rotational arrangement. I'll run the calcs later, but would like to see a second set to verify my work. This may have a big impact bearing selection.

James, I'm not trying to design a new bearing for this application, so I have not run any calculations. I have no interest in providing any competition for LN, IMS Solution, Flat 6, or TuneRS. I am just searching, exploring, discussing, and even arguing about what is the best fix. It's difficult though because there is money involved here and some people are pushing their products with pretty hard sales methods. You have to read what is out there pretty carefully.

For example, IMS Solution has this statement on their website:

"By replacing the factory sealed ball bearing with a pressure fed oil-lubricated plain bearing, the IMS Solution eliminates potential damaged caused by foreign object debris, eliminating 11 wear components from the assembly."

I think this is misleading at best, but more likely deceptive. 11 wear components are eliminated, really? I call baloney on that one in a heartbeat. The bearing is eliminated and the weak shaft that presses into the inner race of the bearing is eliminated. So what are the 11 components then? Well this is an extreme stretch, but I honestly think they are counting the eight individual balls in the bearing as being separate "components", the inner race, the outer race, and then the shaft = 11. When Porsche assembled these engines though they only bought two components from suppliers - the bearing and shaft. Calling that 11 components to try to make the change seem more significant is baloney.

So tread carefully and try to sift through the baloney.

If you want to post your calculations for sizing a bearing, feel free to. I am sure you can get some good feedback. You are limited though simply by what will FIT inside the IMS. If you look at off-the-shelf ceramic bearings be careful about the cage material. Most use a nylon cage. I don't believe the IMS Retrofit does. I believe that is what makes their bearing "custom" is that it probably uses a steel cage or some other metal. I would check the temperature rating for the nylon cages that are standard on ceramic bearings. I don't believe they will hold up in this environment and I believe that is the main factor that stops most folks from just grabbing any old ceramic bearing and selling it as an IMS fix.

Kirk Bristol

[V-Rod]

Kirk,

While I have enjoyed this thread and the depth of the discussion. I do feel that your few use of words discount this thread and your your participation.

[Kirk]

Originally Posted by cnavarro

However, I will clarify that just going to a ceramic hybrid doesn't give you a higher load capacity than an equivalent conventional bearing but their benefits certainly outweigh the cost.

Hybrid Ceramic Bearings | Applied.com

Thanks Charles, the information you provided is very helpful!!!

I do have one question that has been bugging me. Looking at the Applied website that you provided you will see that they recommend grease first and foremost for ceramic bearings, but that's not practical in our application as the grease would need to be changed very regularly. In that situation they recommend oil lubrication and they note that a ceramic bearing does not need as much oil as a steel bearing. However, lubrication is still key. The LN Retrofit relies on oil bath and oil splash lubrication, which Applied notes is fine in some applications. However, they also note that oil jet and circulating oil are also acceptable.

So what if someone does not feel comfortable with the LN Retrofit approach? What if they think that oil bath/splash in this application of high RPM's and high heat is marginal and that a more consistent oil flow is better? What would be the problem with a direct oil feed to a bearing like the LN Retrofit? The LN Retrofit website discourages this practice:

"The LN Engineering IMS Retrofit kit should not be used with any forced oiling products for which they were not designed for."

But ceramic bearings in general are designed for all forms of oil lubrication. Since you are providing some technical detail regarding your ceramic bearing would you mind providing some better detail on why the LN Retrofit does not work with direct oil feed?

Kirk Bristol

[Kirk]

Originally Posted by V-Rod

Kirk,

While I have enjoyed this thread and the depth of the discussion. I do feel that your few use of words discount this thread and your your participation.

Sorry if I've offended. I've edited my post to tone it down. I get too passionate when it comes to engineering and technical discussions. I just love this stuff!

Kirk

[V-Rod]

Originally Posted by Kirk

Sorry if I've offended. I've edited my post to tone it down.

Kirk

All is good...