J-Rod

I have seen the threads flying about, I'm trying to run xxx and I broke a spring. Ok, so lets look at why.

The most common mistake I see in engine building is to use valve springs with inadequate pressure. Not all springs are created equal; just because a set of coils is described as "roller springs" in a catalog or advertisement does not mean that the springs will produce enough pressure to do their job.

There are several misconceptions about valve springs that influence racers to make poor decisions. A customer who says, "I don't need good springs because I'm running stock valves," is badly mistaken. Steel valves are heavy, and adequate spring pressure is absolutely essential to control their motion. A valve's inertia increases with the square of the engine speed, so even a small increase in rpm requires significantly more spring pressure to maintain valvetrain stability.

It is a myth that stiff springs will pop the heads off valves or cause valve tuliping. The only time that the valve head is subject to spring tension is when the valve is closed and resting on its seat. At all other times, the valve sees only a compressive load between the tip of the valve stem and the groove for the valve locks. In our Pro Stock engines, we use 7-inch long titanium valves with tiny 7-millimeter stems and springs that exert more than 1,000 pounds of open pressure - and we've never broken or tuliped a valve due to high spring pressure.

In fact, too little spring pressure is almost always the root cause of valvetrain failures. We spent a year studying valve springs using an Optron, a sophisticated electronic device that can precisely record valve motion and reveal valve float. We learned some shocking truths about valvetrain behavior at high rpm. Even with a relatively mild camshaft profile, the valves bounce on their seats before they close. If the spring is too light, the valve bounces uncontrollably. The valve hits the seat, rebounds, hangs in the chamber awhile, and thebounces erratically several more times. Imagine how hard this is on the valve and the rest of the valvetrain!

Even with high-pressure springs, the valves still bounce when they close. The crucial difference is that the bounce is controlled and predictable, like dropping a basketball. The valve bounce diminishes progressively, and generally on the third bounce the valve stays closed until the next cycle.

The evidence is unmistakable when we tear down an engine that has been run with weak springs: The valve seats are usually beaten up, the valve job is wiped out, and there is fretting on the valve faces. It's fortunate when we catch these problems early because weak springs will almost certainly cause a catastrophic failure.

Another excuse I've heard for not using stiff valve springs is that they take more horsepower to compress. My reply is that each spring stores energy, and for every valve that is opening another one is closing. Anyone who has been whacked by a torque wrench while turning a crankshaft can testify that the valve springs exert considerable force on the closing ramps!

I have never installed stiffer valve springs on an engine and lost power; the improvement in valvetrain dynamics more than offsets whatever additional power is required to overcome the springs' resistance.

I think that any serious big-block drag racing engine should have at least 220 pounds of seat pressure after it has been run. While 1.550-inch diameter chrome-silicon springs may have adequate pressure when they are first installed, they eventually fatigue and lose their tension. We use 1.625-inch diameter Vasco Jet springs exclusively on our Super Series engines and big-block cylinder head packages. These are the same springs that we used in Pro Stock engines just a few years ago; we have virtually eliminated valvetrain breakage in our bracket racing engines by installing these high-pressure springs.

For my money, peace of mind is worth the cost of premium valve springs in any drag racing engine.


Now, you may be telling yourself what does a big block have to do with me. The fact of the matter is that whether it is an SBC,BBC / GEN whatever
the fact of the matter is valvetrain stability is the key. Now, here is some data on the 918's from David Vizrd that I am posting for your reading pleasure. It will help you understand what the 918 is all about...


Suspense might be fine if you are Alfred Hitchcock but I really don’t feel inclined to wait until the end of this feature to tell you how effective Comp Cams’ new (well, relatively new) beehive springs are. Unless some other factor, like an overly soft (and consequently collapsing) lifter plays into the game, these springs, with less seat and nose force and, on their worst day, are equivalent to replacing heavy steel valves with lightweight titanium (Ti) valves. Equally good news is that this is all achievable without the exotic price tag that Ti valves carry.


Although deserving of credit for its introduction to the performance world, let me make it clear that Comp Cams did not invent the beehive spring. During the late ’70s Detroit spent millions to research and cost-effectively produce hundreds of millions of beehive springs. Although far short of a new concept, the beehive spring has only just become a viable proposition for the racing fraternity.

Although it looks simple enough the design of a beehive spring is such that it calls for engine builders to do a complete rethink in terms of spring poundages. To be able to make such revisions in conventional valve spring poundage without a leap of faith means understanding why a beehive shaped spring is so much better.


Spring Surge
So how" you may ask, "is this minor miracle in valve train dynamics achieved?" To better understand how Comp’s spring is so much better we need to first look at a conventional spring’s #1 problem – spring surge. All springs have a propensity to vibrate at a certain frequency – a little like a guitar string. This is known as the spring’s natural resonant frequency.

The principle factors affecting the frequency of the vibrations are the spring’s mass and its stiffness. If the spring is excited at that frequency by the valve train the vibration will rapidly build up to such an extent that the spring is unable to control its own motion let alone that of the valve train it is supposed to be controlling (Figure 1). Dual springs, each with a distinctly different natural resonant frequency and rubbing together during opening are intended (as are flat wound dampers) to damp and counter unwanted spring surge. At the end of the day the traditional use of friction between spring coils or dampers can generate a great deal of heat. Putting some numbers to that should illustrate the magnitude of the situation, at least in the extreme case of a ProStock valve train. While doing some valve train spin testing I measured the temperature rise on such a conventional, steel sprung, valve train and each spring generated in excess of 3,000 watts of heat. Admittedly a Winston Cup motor has a less extreme valve train, but the mileage it is called upon to do is high. The only way these springs can live is to be oil cooled.


Spring surge is the major driving force behind the adoption of air springs on those 19,000 rpm F1 engines. If a super spring is possible then the primary goal is to find a way of eliminating surge without the need to introduce the ultimately destructive element of friction damping into the equation. The beehive spring does just that.

Defeating Mass Effects
With a conventional spring we find that, other than the top and bottom coil, all the coils are evenly spaced. When the valve opens the top coil moves the same amount as the retainer/valve while the bottom coil remains stationary. Checking the motion of individual coils between reveals each moves proportionally less than the one above it. This means that during the opening event the rocker is not accelerating the entire mass of the spring but only a proportion of it. Determining the equivalent proportion being moved through the entire valve lift involves some relatively complex mathematical calculations. However, most textbooks on the subject indicate that the effective spring mass, as seen by the rocker, is typically one-third of the spring’s total mass.

An example here would mean that the mass of a typical 120-gram spring reacts as if it were 40 grams. During the opening event the coils get closer together but, even at full lift, they should not touch. If the coils do clash due to surge the stresses in the spring can spike to two and three times that induced during normal operation and failure will result.

The mode of operation of a beehive spring is somewhat different to that of its conventional counterpart. From the photos you can see the beehive spring’s coils become progressively smaller from the base to the top. For a given wire diameter the smaller a spring’s coil diameter gets the stiffer it becomes. Because of this the beehive spring’s top coils are not only smaller and lighter, but also stiffer.


When installed the coils nearest the spring seat – being less stiff – compress more than those at the top. As the spring is compressed during the opening of the valve, each coil progressively settles down onto the one below. As soon as any part of a coil has settled on the neighboring coil beneath, it is removed from the equation as far as mass in motion and spring rate is concerned.

By the time a valve controlled by a beehive spring reaches full lift only the small top coil or two is actually being moved and the springs stiffness has escalated considerably. Under these circumstances the remaining active coil or coils are very stiff in relation to the moving mass involved. In basic terms this means as the valve approached full lift the effective spring mass moved by the rocker is as little as 10 percent of the spring’s total mass.

Because a beehive spring uses the mass of the wire from which it is made far more effectively it can be considerably lighter. In terms of mass being controlled at or near full lift, this means 10 percent of a light spring as opposed to 30 percent of a much heavier spring. Assuming a 72-gram beehive spring the amount of mass involved at full lift would be about 7 grams. A conventional spring that would allow the same rpm would, at the very minimum, be 120 grams thus having an effective mass, as seen by the rocker, of 40 grams. The difference right here is almost that of a steel versus a TI valve.

So far so good – now let us throw in the effect of the retainer. Because the top diameter of the beehive spring is so much smaller than a conventional spring, and no secondary step for an inner second spring is needed, the retainer weight drops by a little over half. This brings the retainer weight from about 34 grams to 16-17 grams, making a cheap steel retainer for a beehive spring a gram or two lighter than a titanium retainer for a conventional spring.

This brings the over-the-nose weight savings for the beehive springs up to over 50 grams, which is greater than the typical difference between a steel and TI valve! If the budget allows and you want to go with TI retainers then, you’ll find they are a little cheaper to produce than their regular spring counterpart (because they are smaller, there is less material expense).

Even if these facts were the total extent of the beehive spring’s advantage it would be a winner. However, there are still more advantages.

Defeating Surge
The point has already been made that spring surge is a conventional spring’s #1 problem and it can be insidious in as much as it can (and usually does) occur at some point in the rpm range other than at maximum valve train rpm. Without a dyno to evaluate the power curves surge can go undiscovered.


Generally speaking the ideal spring is one that has a very high natural resonant frequency. The higher the natural resonant frequency is the less likely the system is to fall foul of valve train frequencies with the potential to drive the system into surge. A conventional parallel wound spring has a fixed natural resonant frequency value. To raise the frequency the spring must become stiffer for the mass involved or the mass must be reduced for the stiffness involved.

Let’s see how this relates to a beehive spring. First, starting from the bottom and moving up, we find each coil is stiffer and lighter than the last. As each coil compresses the amount of mass being moved is reduced and the stiffness of the remaining operative part of the spring increases. The action of progressively eliminating coils as the lift increases means that the spring has no single defined natural resonant frequency. It is continually changing as the valve progresses through the opening event. This has two major consequences. First, the spring needs no form of heat generating friction damping and second, the spring can use far more of its delivered force to control the valve train rather than itself.


My own experience to date indicates that a beehive spring of about two-thirds the mass of a conventional spring will run a typical street roller valve train to 2-300 rpm more while the seat load drops from 145 to 130 lbs. and over the nose from 345 to 310 lbs.

Optimized Wire Form
It is not commonly realized but round wire is not the best shape for spring wire. Its predominant use is simply a manufacturing convenience. Having already gone out on a limb to bring an advanced spring to market Comp decided to go with a more optimal wire form. The round wire form most springs are made of develops higher operating stresses on the ID of the spring than the OD. This sets the operating limit to whatever the ID of the spring can tolerate. By adopting an ovate wire section stresses are more evenly distributed. This allows the spring to have a little less mass, a shorter nested height for higher lift applications and greater reliability.


Everything so far discussed means that, for a given rpm, a beehive spring can get the job done with at least 25 percent less over-the-nose poundage than a conventional spring. This is far greater than the margin between life and death for a flat tappet cam. Although I have not personally tried it, I suspect that beehive springs can handle a flat tappet cam spring with as much as .600˝ lift to over 7,000 rpm with spring loads that will allow a ball pivot fulcrum to survive semi indefinitely.


The reduced load also means that a hydraulic cam will suffer less collapse than would otherwise be the case. Although this may not be a really big issue with a flat tappet hydraulic cam it is really critical with the far greater collapse and consequently power-robbing problem a hydraulic roller has. Don’t think for one moment I am talking about the odd few HP here as a result of using the currently available spring sensitive hydraulic rollers. By going to a solid quiet ramp street roller cam of less duration I have seen, on a BB Chevy, a gain of over 90 hp and 120 ft.lbs. of torque. Any time an avenue to reduce the spring loads used on a hydraulic roller presents itself it should be pursued as it will pay power dividends.


Currently Comp Cams only has three beehive springs but they are proving so versatile that there is a high probability that one will suit your needs. In fact any one of Comps so far limited range of beehive springs will replace a substantial number of regular spring part numbers so they should reduce what engine builders need to carry on the shelf. This is especially so for the 26918 spring. With appropriate shimming, its seat and nose poundage can be made to handle almost every flat tappet hydraulic cam requirement for any small block domestic V8. When choosing be aware that to achieve the same rpm a beehive spring can be run with substantially less seat and nose force. At this as yet early stage of events I estimate it is possible to drop the over-the-nose poundage by about 25 percent and the seat preload by at least 5 percent to maybe 10 percent for the same rpm potential, so bear that in mind when setting up a valve train.

Source:
Competition Cams
3406 Democrat Road
Memphis, TN 38118
901-795-2400; www.camhelp.com.


Effective Valve Mass Comparison


Steel Valve 125 g

Reg Spring 120 g

Steel Retainer 34 g

Steel Keepers 6 g

Effective Valve mass 205 g




--------------------------------------------------------------------------------

TI Valve 80 g

Reg Spring 120 g

Steel Retainer 34 g

Steel Keepers 6 g

Effective Valve mass 160 g




--------------------------------------------------------------------------------

Steel Valve 125 g

Beehive Spring 72 g

Steel Retainer 17 g

Steel Keepers 6 g

Effective Valve mass 154 g




--------------------------------------------------------------------------------

Steel Valve 125 g

Beehive Spring 72 g

TI Retainer 8 g

Steel Keepers 6 g

Effective Valve mass 145 g




--------------------------------------------------------------------------------

TI Valve 80 g

Beehive Spring 72 g

TI Retainer 8 g

TI Keepers 3 g

Effective Valve mass 98 g







Currently Available Springs


Part # 26915
Seat Lbs. 105 @ 1.8"
Open Lbs. 293 @ 1.8"
Rate (Ave) 313 lbs./in. 313 lbs./in.
Coil Bind 1.085"
O.D. 1.290/1.060"
I.D. 0.885/0.656"

Part # 26918
Seat Lbs. 130 @ 1.8"
Open Lbs. 318 @ 1.2"
Rate (Ave) 313 lbs./in.
Coil Bind 1.085"
O.D. 1.290/1.060"
I.D. 0.885/0.656"

Part # 26095
Seat Lbs. 180 @ 1.9"
Open Lbs. 4308 @ 1.2"
Rate (Ave) 357 lbs./in.
Coil Bind 1.170"
O.D. 1.600/1.185"
I.D. 1.145/1.030"



Now, you may be asking yourself, what does this all have to do with me? The fact of the matter is we are right at the limits of what we can handle currently given the choices folks are making. So, you may ask yourself, well, what can I do? The choices become pretty slim. You can upgrade to a a comp 977 with a Comp R lifter with .010 pre-load. You can run 918's or some equivilant spring with the stock lifters and run a risk of breaking the springs.

So, you may be asking well, how can I have my cake and eat it to. The simple answer, and its really not so simple, is to lighten up the valvetrain even further.

Here is where I think you could make an LSx valvetrain really work.

Use the cam of your choice with a 918, use a Ti retainer, and use one of two things either a NaK filled valve like the Z06 which makes the steel valve as light as a Ti valve, or switch to Ti.

I know, Ti is expensive, but until you start reducing valvetrain mass you are going to walk this line between too much spring vs collapsed lifter vs valve float/broken springs...

Just my $.02

(Wondering out loud here) I wonder if we could get the Z06 Valves in a 2.02-2.08 size. I wonder who mfgs them for GM?

gomer

iTrader: (13)

I posted a link to that article a couple of weeks ago.. back when I was debating on my valvespring choice. I ended up doing the 918's with the Ti retainers based on that article and other information. If I would have used a big (230+) cam with XE-R lobes.. I wouldn't have been comfortable doing it.

J-Rod

I think if you went with a Ti (or equivilant) valve you could go with one of the big cams with the 918.

CHRISPY

I think Comp R lifter is pretty important to consider in this. ESPECIALLY on XE-R lobes.

Cheers,
Chris

gomer

iTrader: (13)

I had considered going with a Ti valve.. but I couldn't find anything that seemed to be within a reasonable cost. Is anyone aware of a Ti valve that will work in our heads at a "decent" price?

TVWilkes

I have a few sources for them.Ti valves costs around $80.00 each.I will find a "decent" price.

Camaroholic

iTrader: (1)

When Geoff at Thunder was building my motor, before they put it together, he spent some time on the Spintron looking at valve bounce with different valvetrain combos. He gave me some very interesting graphs (and he hung on to a few of them from Jason's old valvetrain). If you look at Jason's (solid roller motor) dyno chart:

http://www.thunderracing.com/images/jason_460dyno.jpg

Look at how it craters around 7000 RPM. That's valve bounce. Geoff saw it, and measured it, on the Spintron. He was getting severe bounce. He had harmonics problems. FWIW, that's what destroyed the heads.

My motor, which was assembled using a slightly less aggressive cam profile (a bigger cam in size, but different lobe profile) and different springs, doesn't have near the falloff that Jason's had...

http://www.akmcables.com/trdyno.jpg

Now, the spintron did show that my valvetrain becomes unstable at about 8000 RPM... But I won't be spinning that high.

I might be able to dig up the Spintron graphs if anyone's interested in seeing what one looks like...

And I hope to have new dyno numbers on my motor in a few weeks. We're making some significant changes.

Andrew

J-Rod

I agree that a Comp R lifter is important since you have to go up to a 977 to get the valve control you need. But, if you can reduce mass then a 918 would be enough to get the job done, and hence still work with the stock lifter. Of course, there is something to be said for the reduced ammount of lash in a CompR.

KGSloan

this is all very interesting. thanks for sharing your knowledge with us j-rod

SScam68

iTrader: (20)

Camaroholic

This would be a great post to put those graphs in. It will also compliment what J-Rod posted above.

Anybody know how much the ZO6 valves cost? Terry?

I wonder why comp, or other manufacturers for that matter don't give a suggested valvetrain mass or rpm window. Not that simple?

Ragtop 99

iTrader: (7)

What does the Comp R lifter add compared to stock? I'm running the double springs on stick lifters now.

MyLS1Hauls

iTrader: (2)

They are more durable and won't pump up at high rpm.

66ImpalaLT1

I think only the Z06 exhaust valves are filled with NaK.

VINCE

iTrader: (1)

What about stock valves? Are they too heavy. I am talking about 01 #241 heads. I am running a 232/236 .59x/59x lsa 114.. My car shows no sign of spring fatigue..

Cal

iTrader: (1)

Excellent information! I already have Ti retainers, but didn't know you could get Ti keepers also . . . anybody know how much they cost or where to get some?

Scalpel

iTrader: (4)

I can get the NA filled intake/exhaust valves from the LS6, new for $45 each

J-Rod

NaK filled valves in a biiger size would be nice though. I agree that the stock Z06 valves are light. It'd be nice to try and line up some larger valves, or some Ti valves to lighten the valvetrain mass.

BTW thanks for the graphs...

PacerX

iTrader: (4)

There are a few things I really wish for in this life:

First being a properly designed spring, spec'd out to the point of...

a. Maximum valvetrain mass
b. Maximum rpm for a given valvetrain mass
c. Most aggressive lobe profile possible
d. Maximum lift

Then you could have a (fairly) simple chart...

If your valvetrain weighs "x", then the maximum lift becomes "y", your max reliable rpm becomes "z" using lobe profile "a" at lift "b".

This is the type of issue that bothers the hell out of me. If ANYONE actually did their homework while designing springs, they HAD TO go through this process.

Otherwise, they're just shooting in the dark.

Sadly, most of the aftermarket shoots in the dark.


Just like pushrods. There ought to be a dmaned chart out there that says:

Deck the head .010, go to a .020 smaller base circle cam, with stock lifters and stock thickness gaskets and "Brand "x" rockers and your correct pushrod length is "y".

Holy hell. Somebody get GM to give me the CAD model and I'll calculate it out.

Raughammer

Man that is a good article1
While i agree with most of what has been said. I have successfully used and seen others in my core circle of buds use the stock lifters on some pretty radicle cams.
I have NEVER used anything but stock ones on my motors and have not had any valve train issues to speak of. (one bad spring after 15 minutes of run time...it WAS a bad spring)
The stock lifters are good to go for 99% of most folks applications on here.
Thats just my opinion and has been my experiance.

Again.. thanks for posting that VERY informative article.

J-Rod

I think you can use a stock lifter in any application providede that the mass of the valvetrain is such that you do not have to have crazy spring pressure to keep the valves under control.

Thats my whole deal, I think you can use a stock lifter, and a 918 with any of the cams currently out (even the really big ones) if the rest of the valvetrain is light enough. Say a Ti retainer, and either a NaK filled steel valve or a Ti valve...