BBC solid roller lifter failures - a root cause analysis

[PackardV8]

The famous 30-30 (thous) Duntov cam came into existence when Corvette Chief Engineer Zora Arkus-Duntov found that the cause of burned exhaust valves in racing (all iron) SBCs of the era was the then specified ~0.010" cold exhaust lash going negative at full operating temperature, due to expansion of the 1400°F. exhaust valve...

Seldom do I question Bill on anything he says here. However, it would be good to see a reputable reference, as I've heard a different auto/urban legend on this one. Ed Iskenderian once told me Ed Winfield said he designed the cam for Duntov. Winfield had gotten reclusive in his later years, just as Duntov had hit his stride as self-promoted public face of the Corvette. GM was slow to react to Duntov's wanting to move Corvette into racing, so he went outside to Winfield for a quick fix. The story goes, Winfield was a self-taught genius, but at the end of his career, hadn't really made the transition from flatheads to the SBC and the .030" was a result of the effect of the 1.5 rocker ratio on some of the transposition of his flathead grinds.

Either way, few subsequent cam profiles required such a large running clearance.

thnx, jack vines

[CamKing]

The story goes, Winfield was a self-taught genius, but at the end of his career, hadn't really made the transition from flatheads to the SBC and the .030"

Winfield designed the cams for all the offy's my dad built from the 50's, into the 70's. He also designed plenty of other types of cams for my dad.
When my dad started developing his cam design formula, he ran everything he did past Ed for approval.
The masters of Winfield's designs that I have, have ramps in the .012"-.015" range.

[Lasher...]

Chrysler was one of the 1st to really expirment with wide lash mechanical cams.... they settled on .028/.032 cold on their Max Wedge then the Street Hemi's .... Chevy jumped on the bandwagon with the 425HP 409's and then the 30-30 Duntov in 1964.

As I understand it burned exhaust valves was the major reason for trying the wider settings........ And under extreme loads and temps the wider settings worked..... Ford went to .025 on thier 427's also.....then the 396 came out with .024/.028 hot........... It seems the .012/.018 type lashs common back then worked fine for street use but under heavy loads and heat of the racing engines it just wasnt enough with the weak cooling systems we had back then as well as the different exh valve materials that were common then compared to now.

By 1967 I have seen BBC flat tappets with lashes as wide as .036.... yes .036. Cam Dynamics on one family of their roller lobes used .035 / .030.on SBC lobes.

Yet for normal street use the mopar slant 6 used .010/.020 and often would go decades between needing to lash the valves.

Under the most extreme heat and loads the wide lash does offer some extra margin of clearence so if the valves did grow then the valve will still seat when the lifter is on the base circle.....Does it need .036 .... most likely not.....at least not with todays superior products.

[540 RAT]

Who actually made the decision to incorporate an average of .024-.028 for a mechanical roller camshaft?

Well, a lot of high performance and even current OE cam design uses long constant velocity clearance ramps, and have done so from the '30's.
Both Mike Jones and Harold Brookshire have devoted many posts to the subject of lash.
In the context of roller loading and failure, the important factor is not the lash, but what velocity is encountered at the lash point. The distance does not matter if the velocity is low. If however, the cam designer picked a "safe" clearance point say at .020", and the lash is opened up to .026", the collision speed is increased dramatically and loading goes way up.

None of the above has to be and there is no accurate data that proves this to be benificial.

When this discussion started, I pulled out some old "good used" roller lifters that I had kept around for spares, and pressing out the center pins I observed the surface fatigue that 540 RAT was talking about. I have never tried to look at this before. The needle rollers still look good, and the inner and outer surfaces of the roller still look serviceable, but the support pin has eroded clearly around the bottom third of the pin and eventually would lead to a failure. So much for "good used", NOT.

Edited

Thanks for the input about the "good used" lifters that you took apart. Then ones I took apart had already gone far enough that most or all of the needles showed failure as well as the axle. So it was somewhat inconclusive as to which went first, though the axle was the worst off in each case, suggesting it went first. But the axle is softer so it was a little hard to say which came first, the chicken or the egg. Your lifter examination would certainly seem to confirm that the axle went first, as suspected. Thanks for letting us know.

[540 RAT]

Something else that is probably worth adding to the discussion is that, I reviewed a study on diesel lifter failures, which found that their roller lifter failures happen earlier or are much more frequent when the oil is too slick, which they called "low traction oil", which you can find with some of the premium synthetics or low friction racing oil. And their data showed that the lifter failures were due to the roller skidding, sliding, or skipping on the lobe. Also thicker oil contributes to skidding as well. This skidding significantly reduces the fatique life of the metal the lifter is made of. And they found that "high traction oil" as they called it, reduced the failures significantly. An oil I found that fits this "high traction" definition nicely, is Valvoline 4-Stroke Motorcycle Oil (dino type), which is available in 10W40 and 20W50 at Autozone and some other Auto Parts Stores. It is optimized to not be too slick for the wet clutches that motorcycles typically run. A buddy of mine runs it in his 9 second Japanese motorcycle Sport Bike, and likes it just fine. Of course most bikes make more hp per cube than our stuff ever will, and they turn more rpm than our stuff ever will (his stock redline is 14,000 rpm and that is not a typo), so using it in our stuff should be a walk in the park. Since many of us use slick synthetic or racing 20W50 oil, this is something else we can consider.

[TheEngineWorks.com]

I don't work closely w/ any Pro Stock stuff, but I know Matt Hartford from Total Seal who was driving Nickens NHRA Pro Stock earlier this year for about 9 races, and I am gonna ask him some more details about the lash they run. As I understand it, some of them run the lash down to about nothing cold. I'm sure some of you could pipe in here with Pro Stock experience and give some details.

Seems to me that lash is really just a function to keep the valves on the seat during extreme conditions, although with ANYTHING with aluminum involved I would say that under heavy loads it would be almost impossible to get the lash to close up, even if you ran ZERO lash cold.

[jeffmckc]

Mike Jones did my Drag Week cam this year, and its a tight lash cam for just this reason Edwardo.

[CamKing]

Seems to me that lash is really just a function to keep the valves on the seat during extreme conditions, although with ANYTHING with aluminum involved I would say that under heavy loads it would be almost impossible to get the lash to close up, even if you ran ZERO lash cold.

If you run tight lash on a cam that's not designed for tight lash, you'll be increasing your seat duration, and that will hurt your performance.
If the cam has constant velocity clearance ramps, tightening the lash will not reduce the hit the lifter will take.

[stix818]

So your thinking possibly by going to a lighter oil this would help eliminate the skipping of the roller?

I'm curious why you guys use a thick oil like 20w-50 now? I run 30w redline racing in my dragboat and I thought that was on the high end of the viscosity range for a drag motor. I've read where the pro-stockers (bikes and cars) are actually running 0-2w which is pretty interesting in itself!!

[MadBill]

...Seldom do I question Bill on anything he says here. However, it would be good to see a reputable reference, as I've heard a different auto/urban legend on this one. Ed Iskenderian once told me Ed Winfield said he designed the cam for Duntov...thnx, jack vines

Well, I got my pay check from The General for 35 years Jack, so it behooved me to accept the party line... I guess if Duntov cut the check to pay for it, he got to put his name on it.

BTW, this cam was used up until the '70 Corvette /'70 1/2 Camaro LT-1 but only with the manual trans. The Camaro Z28 became available with an automatic staring that year and the cam had to be detuned to accommodate it.
The milder cam actually made more power through the stock exhaust than did the Duntov, but significantly less with open headers.

[Rommel Racing]

If you run tight lash on a cam that's not designed for tight lash, you'll be increasing your seat duration, and that will hurt your performance.
If the cam has constant velocity clearance ramps, tightening the lash will not reduce the hit the lifter will take.

Mike,

Would you know what kind of clearance ramps my Comp cam has? I am running their Hi-tech .440 lobes, #4078 on intake and #4080 on exhaust.

Sorry I have to ask,but Comps tech guys wouldn't likely know

I also run an all aluminum bbc on the street,and have been setting the lash cold at .008" and getting .014" growth hot. It seems to run fine like this,but I would like to run it a little tighter if possible.

Thanks,
Steve

p.s. knock on wood,but my Isky EZX rollers seem to be working really well so far

[MadBill]

How many miles to date, Steve?

[Wolfplace]

How many miles to date, Steve?

Hi Bill,
Not to answer for Steve but I have a couple of sets with pretty fair street time on them now which I mentioned in an earlier thread about these lifters.
One has probably 8K & the other set has about 15K now last I heard & so far so good
These are the EZ not the EZx
I also have a number of other sets out there but do not have any feedback timewise which is most likely a good sign

A number of folks joined in with mostly positive results & probably the best endorsement for endurance came from Bob Madera
You can look through it all here.
Bob's comments are towards the end
http://www.speedtalk.com/forum/viewtopi ... rs&start=0

[jimivice]

I read your analysis and it is based on the axle. Is the axle being used as a race. It is my opion that roller should be a captured bearing. Meaning that the axle should be used to retain the bearing to the lifter body-not as a race. In the past people would look at the pin, or axle diameter, and think it is too small. They related the pin diameter as being small and could not hold up under the load. A market was created. Manufacturers made the pin diameters larger and used the pin as a race and did away with the inner race.. The lifters failed and then they offered lifters with oilers to direct oil to the roller. The premise of roller lifters was to operate with minimal oil and now they have to be supplied with oil. I agree with, that with the spring loads of today the lifter is more taxed. Is it because of bearing failure or a poor design. It is my opinion that lifter is designed to fail. There is no way rolllers can rotate about the axle without wearing out the axle or pin- no matter how large you make the pin diameter or oil it. It must be a captureed bearing with a seperate inner race. Schubeck's selling feature was if the convential roller failed, you would have the failed needle bearings being pumped through the motor and consequently destroying the motor. He apparently so a flaw and offered an alternative. I saw his lifter at a PRI show and he offered a viable alternative. I'm not an engineer but I rememember when roller lifters were the answer to flat tappet cams and now they have become questioned. It does not make sense that a basic design, by an engineering stand point, is now questioned.

[kirkwoodken]

I posted this reply earlier today on the Chevelle site about the same subject.

In a sleeve type bearing, such as a rod or a main bearing, if the bearing clearance is .001" on a side, you can have a piece of dirt .0005" in diameter floating around in the oil, and it will never have metal to metal contact.

With sliding bearing, like that of a flat tappet, if a small piece of dirt is slightly thicker than the oil film, it will tend to be ground into smaller pieces that are harmless.

But, in a roller tappet, you virtually have metal rolling on metal. It is the hardness of the materials involved that makes roller bearings work. That is also the reason they fail. If a piece of dirt gets between a rolling surface, where there is direct contact between two surfaces, the dirt has a tendency to go into the surfaces. If you place a human hair on a piece of steel, and roll that piece through a steel rolling mill, the hair will leave an indentation in the piece of steel. Why? Because of the tremendous force concentrated in a small area. Add to the effect of dirt the tendency for a small needle bearing to Brinell the axle and you can see why the rollers may have problems. Everything is just too small for it to last forever.

This is the best way I know how to explain this; maybe someone can do better. The most important thing with roller lifters is super clean oil. No 6000 mile oil changes for highly stressed rollers. I do not have remote double filters yet, but they are coming.

As for how rollers hold up in stockers, I would like to see some info on failure compared to oil change frequency. I think the dealer supplied oil changes may contribute to their longevity.

I do not know what type steel the axles are made from. I think if they were made from something like Carpenter W-1, which I can't remember the trade name for, and cyro quenched, they would be about as hard as something could be on the outside and still have a tough core. Vasco Supreme, which is about 72 Rockwell C, can be cryo quenched, and it will remove its inherent brittleness. Maybe someone would like to try that.
There are also the new diamond like coatings that are being used on wrist pins. And Ti nitrides. And how about TuffTride or would that promote cracking?
I don't know how many surface treatments have been tried to increase the longevity of the roller lifters, but it is something I would find interesting to be a part of.