Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Valve spring in engine, why the closer coil must be down?

  1. Nov 24, 2011 #1
    In engine, there is valve and the valve spring. If you have ever look at the valve spring before, then you will notice that one end has its coil closer together than the other end. So a valve spring is a kind of progressive spring.

    In almost all engine repair manual I read, the instruction is always say that the end of the valve spring which has closer coil should be facing down (i.e. closer to the valve).

    1) Why the closer coil end must be always touching the valve?
    2) What will happen if the spring is put upside down?

    I have search in the internet and I found that the reason for using a progressive spring as valve spring is for the dampening effect, otherwise more complex alternative approach has to be used such as an inner spring design.
    However, the spring as a single entity, wouldn't it performs the same regardless of how we put it? Let's be more general, imagine we have a progressive spring (not necessary the valve spring but a general progressive spring), and then this spring is used to do the normal oscillation experiment. In the first experiment, the end with closer coil is facing down. So we get a set of measurement of amplitude and frequency of oscillation.
    Then repeat the experiment but let the end with closer coil facing up, we get another set of measurement of amplitude and frequency of oscillation.

    3) Will the first experiment result and observation identical with the second experiment? If it is not, why?
    Last edited: Nov 24, 2011
  2. jcsd
  3. Nov 24, 2011 #2

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    Unique shape provides better valve control while reducing weight and lengthening life
    Increased harmonic resistance increases stability and reduces damaging harmonics
    Oval/multi-arc shape places maximum surface area at the point of highest stress
    Revolutionary design yields more stability, higher RPM and more horsepower
    Less valve train reciprocating mass in motio

    Attached Files:

  4. Nov 24, 2011 #3
    These are more like the function of the uniquely shaped valve spring, but it doesn't explain why the more coil end has to always face down.

    BTW the spring picture you provides, doesn't look like a typical valve spring of engine, maybe those that I seen are old design.
    The typical valve spring look like this:


    Last edited by a moderator: Apr 14, 2017
  5. Nov 24, 2011 #4

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    Thanks for the photo
    the shape provides better valve control with Increased harmonic resistance. This increases stability and reduces damaging harmonics.
    The close coil design places maximum surface area at the point of highest stress

    In short , this design reduces potential of valve float when engine is over revving
  6. Nov 24, 2011 #5


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    I think the OP is making the point that if we were to assume the spring had negligible mass, there wouldn't be any reason to have the spring installed one way versus the other. If we looked at the spring and valve assembly under static conditions, then for any valve lift, the spring load on the valve is the same regardless of which way up we install the spring. The valve spring doesn't 'know' that it is upside down in the static condition or if we assume the spring has negligible mass. So the assumption that the spring is static or has negligible mass must be wrong, but why?

    The difference is that one end of the spring is static and the other end of the spring is dynamic. The end of the spring against the cylinder head isn't moving but the end of the spring against the valve spring retainer is moving very quickly. Note that if we were to determine the velocity of any part of the spring from one end to the other, there would be a roughly linear change in velocity from 0 against the valve head to some maximum velocity at the spring retainer. The coil velocity increases the closer we get to the spring retainer. The objective is to minimize the amount of weight on the spring retainer during this motion.

    In a progressively wound spring, the more closely wound coils make contact after a small amount of movement and become ineffective. That's what raises the spring constant. The spring constant increases for a progressively wound spring when these coils make contact and no longer flex. If they aren't contributing to the spring load, it doesn't make sense to have them moving up and down. That just adds weight to the valve assembly. So the point of putting them against the cylinder valve head is to minimize the weight that is moving (the sprung weight is being minimized). When you do that, the resonant frequency of the assembly is increased - there is less mass being supported for any given spring load. And for a spring that's moving as fast as these springs do, that makes a difference to the natural frequency of the assembly. You don't want the natural frequency of the valve assembly and the frequency of the motion to coincide. That explanation should coincide with what Ranger Mike is saying, but hopefully it helps explain in a bit more detail.
    Last edited: Nov 24, 2011
  7. Nov 24, 2011 #6
    Thanks to Ranger Mike and Q_Goest.
    Q_Goest you correctly interpreted my intention. Your explanation helps a lot, it now make a lot more senses to me. Let's see will there be any other suggestion on the purpose of the orientation of the valve spring.
    Nevertheless, Ranger Mike provided enlightening information too.

    So I believe question (1) has been answered. Would you mind to comment on the question (2) and (3)?
  8. Nov 24, 2011 #7


    User Avatar
    Gold Member

    question 2 has been answered, by the answer for question 1.
    Same for question 3.

    In a static situation, the spring in either orientation will perform the same.
    When oscillating, either orientation will perform differently as described above due to one end of the spring having stightly greater mass and due to a variable spring constant as the spring is compressed.
  9. Nov 24, 2011 #8
    After you had pointed out for me, I agree that question (2) had been answered.
    As for question (3),


    I guess the answer you mean is "No, the result and observation are not identical." I agree with this statement for real case. Let's try to bring the analysis to the next level, by comparing the amplitude and frequency of the two experiments. Denote the amplitude and frequency of the 1st experiment (closer coil end facing down) as [itex]A_{1}[/itex] and [itex]f_{1}[/itex], and the same goes for 2nd experiment (closer coil end facing up), [itex]A_{2}[/itex] and [itex]f_{2}[/itex] , then, will you say:
    [itex]A_{1}>A_{2}[/itex] or [itex]A_{1}<A_{2}[/itex]
    [itex]f_{1}>f_{2}[/itex] or [itex]f_{1}<f_{2}[/itex] ?

    For ideal case, the spring has negligible mass, although coil bind still occurs. Q_Goest explains that the problem of real spring occurs due to the added weight that doesn't contribute to spring constant. Now with an ideal spring, this problem shouldn't occur right? So if question (3) is now concern only an ideal progressive spring, will the answer still the same?
  10. Nov 25, 2011 #9

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member


    Progressive springs do not make contact with each other..danger

    Once assembled on the cylinder head a MINIMUM clearance OF .010"
    should be used between EACH spring coil at minimum installed height ( max cam lift).

    Reason to change Valve Springs

    The reasons that a cam would require a change of valve springs are: 1. more valve lift than the OE spring can accommodate before coil bind. 2. A quicker valve opening and closing rate that requires more spring force to keep the valve from floating. Several other factors that dramatically affect valve spring choice are maximum RPM and valve weight. The force produced by the spring must control the kinetic energy of the valve. Kinetic energy is equal to mass times velocity squared. If you increase your rev limit by only 500 RPM, spring force at full lift must increase by approximately 10%. Usually the spring safety is 10% or slightly more but if you changed the cam and also increased the rev limit you may now have created a valve float problem. It is the same when you replace an OE titanium valve with a steel valve. The steel valve weighs 75% more than the titanium valve that the spring was designed for. Your spring must now be 75% stiffer or you must lower your rev limit by 15%!

    The Valve Spring

    A valve spring's job description is pretty cut and dry: It has to store energy so that the valve and its companion hardware can return to the seat. Sounds simple enough, and it is—until engine speeds increase. At higher rpm levels, a spring is taxed. This can eventually lead to valve float.

    Most people envision valve float in a pushrod engine as the valve lifter physically parting company with the camshaft lobe. While that might (and often does) happen in severe situations, the first phase of the problem actually involves the valve bouncing off the seat. This bouncing action is virtually uncontrolled, and the end result is extended (and unintended) periods of valve overlap. Since the valves bounce open and closed during the compression stroke, horsepower is diluted by a dramatic margin. This rapid bouncing action (coupled with the invasion into the compression stroke) typically causes the engine to pop, bang and miss.

    The valve springs perform the following functions;

    Lifting the weight of the valve
    Overcoming friction on the valve shaft when the valve closes Creating enough friction, or drag, to keep the valve train and valve following the camshaft profile accurately, by always providing (slightly) greater force than the inertial force of the accelerated mass of the valve. At the same time, the valve-spring forces must not be so large as to create excessive friction on the camshaft and potential loss of performance.
    The sole purpose of the valve spring is to force the valve to follow the path of motion that is dictated by the camshaft lobe. Excessively stiff springs will steal torque from the engine and increase component wear while a spring that is not stiff enough will allow the valve to “float” or loose contact with the camshaft lobe. Valve float will eventually lead to component failure. The spring must also have enough available compression distance to allow the full amount of valve lift that is produced by the cam. Coil Bind is the term used to describe a condition of too little compression distance where the spring “bottoms out”. Coil bind will also lead to component failure. More on coil bind later..

    The force required to keep the valve closed (for a length L1) is referred to as the preload force F1. The force required to maintain accurate camshaft tracking at high revs (for a length L2) is referred to as the spring force F2.
    With regard to the dynamic behavior of a valve spring, it must be considered that they have a lower resonant frequency than other components, meaning that they can easily be stimulated to undergo resonant vibration. This means that the spring, in the presence of an appropriate stimulus, no longer follows the cam profile exactly, which leads to periodic excess tension and variations in the contact forces - including possible failure - and thereby results in a loss of function which leads to a loss of performance. In the worst case, the excess tension can lead to the spring breaking. It can, however, also be observed when there is no resonance, that at higher RPM, the springs do not follow the cam profile exactly. There are two main effects that must be considered. Firstly there is the additional lift past the cam (known as loft), which leads to excess tension and secondly there is the bounce after the down stroke of the cam pitch, which results in the valve reopening for a short period.

    There are two options for examining the movement of the springs at different RPM. A laser can be used to record the movement of the valve, as this coincides with that of the spring. The second, more precise option is to measure the behavior of the springs using wire-resistance strain gauges affixed to the springs. The gauges change their resistance as they are stretched and relaxed during the oscillation of the springs. The changing resistance is recorded, making it possible to calculate the deformation of the material and therefore the motion of the springs. These methods allow us to design springs with smaller excess tensions, resulting in increased dependability and reduced performance losses.

    Both procedures are used a Valve Train Dyno, which is able to spin at up to 15,000 rpm and which provides crucial information for optimal valve spring design and performance.

    Because the resonant frequency is directly related to the spring rate and the mass of the spring, it is desirable to select as high a rate as possible, thereby increasing the resonant frequency, in order to avoid the occurrence of resonance within the engine´s working revolution range. Since, however, the limits of the stress range must be considered in designing the spring, in certain cases it is not possible to prevent the resonant frequency of the spring from falling within the working RPM range.

    There are many options for ensuring that the springs function correctly across the entire RPM range:

    Shim the stock Spring
    Stock springs float around 4500 rpm, on a stock cam, so it's pretty obvious they won't work with a performance camshaft. Back in the good old days, shade tree mechanics simply installed a handful of shims under the valve springs to increase pressure, but this lead to a host of other problems, which usually resulted in valve train failure.

    Replace Stock spring with a stiffer one

    It's much better to use a Single stiffer spring to avoid any possibility of coil bind.

    Single spring with a damper to reduce damaging harmonics in the valve train. The only drawback to using dampers, is that they usually require machining the valve spring seat.
    The spring damper (if so equipped) is designed to absorb spring vibration. To oversimplify, spring vibration is much like a sound resonance traveling through the valve spring. If this resonance is timed just right, then the valve spring actually loses its effectiveness as a spring. Any time the actual spring is stiffened, then the natural resonance is increased. The spring damper counteracts this phenomenon. Dampers are not for street car use and are not used in production engines, typically because they produce friction and wear by design.

    Typically, a damper will be constructed with flat sides. The sides of the damper fit tightly against the inner portion of the spring. As the valve spring goes through its motion, the damper actually rubs the side of the outer spring. In turn, it "damps" out the resonance or vibration (to a certain degree).

    Damper failure is more common that we'd like to think—especially on high-lift, radical-profile camshafts. Occasionally, a damper will physically "unwind," and the lower portion of the assembly will work its way between two lower coils of the outer spring. Naturally, this stacks the spring into coil bind. When that happens, all kinds of carnage can occur if you don't catch the problem immediately. In most cases, selecting the correct length of damper will suffice, but if the problem plagues your application, it can be solved by slightly shortening the damper. Beyond this, and old racer trick is to glass-bead the damper after it's deburred and chamfered.

    Progressive springs
    For progressive springs, the spring rate increases within the stress range, which also changes the resonant frequency. This means that the individual resonant frequencies are passed through for such a short period that no resonance occurs. However, because of fitting constraints, progressive set ups are not always possible.

    Dual Springs with frictional damping
    Inner and outer springs are used which rub together slightly, providing an engineered damping factor. If one of the two starts to resonate, the other damps out the motion before it can cause damage. This damping factor utilizes controlled friction, which can lead to a reduction in the life span of the springs. The inner spring in a dual-spring package can accomplish almost the same job as a damper. Typically, the inner spring will feature much lighter construction than the outer spring. Because of this, it vibrates or resonates at a much different (higher) rate than the outer spring. With different points of resonance or vibration for the inner and outer springs, then the ultimate RPM potential for the spring is increased over a single spring—even if the dual-spring package has identical open and seat pressures when compared to the single spring. Of course, this is seldom the case: In almost all applications, the added inner spring serves to increase both the open pressure and the seat pressure of the overall spring package.

    Dual Springs with Damper
    A common combination is a dual spring with a damper. Essentially, the use of a damper with a dual spring allows the spring package to deliver more RPM without excessive spring pressure. However, the damper won't account for major gains over and above a common dual-spring setup without a damper. And where the dual valve spring is perfectly matched to the camshaft application, a damper isn't totally necessary. On the other hand, past experience has shown that a dual-spring package complete with a damper provides better spring life over the long haul. Given a choice (and if your application can accept it), use the dual-spring/damper combination.

    Triple Springs
    What about triple springs? For the most part, triples are best used in roller camshaft applications. They work very well in short bursts of extreme RPM, but in many mild applications, a run-of-the-mill double-spring/damper combination will work just as well. Furthermore, the design and construction of an optimized triple-spring package is totally another story in itself.

    Why can't you just slide in the biggest available springs and be done with it? That'll work with a roller camshaft designed for drag racing only, but for use with a flat tappet camshaft of any sort (hydraulic or solid), too much spring can be worse than too little. Typically, a spring with anything more than 335 pounds of pressure on the nose (open pressure) will rapidly increase the wear on a flat tappet camshaft (along with wear in other areas such as cast-iron guides). Depending on the engine design, the cam profile and the valvetrain geometry, the practical limit for open pressure on a flat tappet cam is approximately 375 pounds. Any more and you'll probably be faced with a pile of broken camshaft pieces.
    But there's more to the "too much is just right" scenario. If (and it's a big IF) the engine can physically live with a large amount of open spring pressure, you're still behind the 8 Ball. Why? Large amounts of valve spring pressure can eat horsepower. Increasing the open pressure by 50% also increases the amount of friction inside the engine. Simply stated, heavier springs require more horsepower to move the valvetrain. In the end, oil temperature increases and power levels can drop by five, six or more horsepower.

    There are three primary spring shapes:
    conical (tapered)
    beehive (cylindrical with conical spring part on one side)
    Both conical springs and beehive springs make it possible to reduce the masses being moved - on the one hand simply through the spring´s tapered shape and on the other through the smaller upper spring retainer, which again leads to an improvement in dynamics within the valve train. In comparison with conical springs, beehive springs offer the ability to operate with frictional damping for spring sets in the cylindrical parts of the inner/outer springs.

    Myth- Too much spring pressure is hard on valves – In truth, what’s hard on valves is the speed at which they contact the valve seat when closing. What dictates how hard the valve hits the seat? It’s supposed to be the camshaft closing ramp (shape of the cam lobe), but when the spring pressures are too low, the valve does not follow it’s intended path and instead slams into the seat and actually bounces. Hence higher spring pressures can actually aid the valve by forcing it to more closely follow the shape of the cam lobe. On the other hand, to much pressure adds horsepower robbing friction. It also increases the wear and tear on valve spring components. IE: broken rocker arms and valve springs, bent push rods, worn valve guides, and so on. Therefore it is important to match the spring pressure to the profile of the camshaft. However you also need to take into consideration the intended RPM range of the motor. Basically...... faster ramp speeds (more lift for a given duration), and/or higher rpm's, require increased spring pressures.
    Coil Bind myth -
    Coil Bind is the term used to describe a condition of too little compression distance where the spring “bottoms out”. It also applies ANY TIME coils come in contact with each other during spring compression. It should be avoided at all costs. Contrary to common belief, progressive springs do not make contact with each other during compression.

    Any spring installation should require you to examine the relationship between the inner spring and the damper to both the cylinder head seat and the valve spring retainer. Due to different designs in springs, retainers and spring seats, there might be coil bind at these locations, but no coil bind on the outer spring. Have a close look as the engine is turned through a cycle (by hand). In the case of a poorly selected spring (or spring retainer), don't be surprised if you see coil bind on the inner spring(s). If that's the case, you have to tear everything apart and install an inner spring that suits both the application and the spring retainer.

    If you buzz the power plant on a regular basis and the engine goes into early valve float (or bounce), the spring material can actually become annealed by the heat buildup. The majority of valve springs are heat-tempered at 400 degrees during construction. Valve float can cause the spring to exceed that temperature by a significant margin. In most cases of valve float, the spring gets so hot that it glows red. This increased heat eventually kills the valve spring. This is magnified considerable when you have these hot coils making contact with each other during coil bind. This makes lubrication of the valve train critical but this is another discussion.

    Once assembled on the cylinder head a MINIMUM clearance OF .010"
    should be used between EACH spring coil at minimum installed height ( max cam lift).
    Last edited: Nov 25, 2011
  11. Nov 25, 2011 #10


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi Ranger Mike,
    You seem to be saying that the coils of progressive springs don't make contact during operation. Is that right? If they don't make contact, they aren't being used as progressive springs, which would beg the question, why they would be wound as progressive springs?

    Ref: http://www.tuninglinx.com/html/suspension-springs.html

    Edit: RM, I know you've had a number of very good posts regarding motor vehicles and especially racing, so I don't dismiss your comments without consideration. Also, to help explain what the issue is, I'll try and explain a little deeper. Looking at valve harmonics, there are 2 basic areas I see as being applicable here, 1) the frequency of the valve assembly and 2) the frequency of the spring itself. These two are different phenomena and they don't in general coincide.

    1) The frequency of the valve assembly is a function of the mass and spring constant. A spring, whether it is a progressively wound or a conventional wound spring, will have a frequency given by the mass of the sprung assembly and the spring constant as shown for example by the Wikipedia article: http://en.wikipedia.org/wiki/Simple_harmonic_motion
    Here, the natural frequency is a linear function of the spring rate, so if the spring rate does not change, the natural frequency of the sprung mass will not change, nor will any secondary frequencies change. So a progressively wound spring that doesn't have coils that come into contact because the spring isn't being compressed far enough, will have the same harmonic frequencies as a mass with a conventional spring.

    2) The frequency of the spring itself is a linear function of the number of active coils as shown for example by efunda:
    So again, the frequency of the spring won't change unless the number of active coils changes.

    So far, we have no way of changing any of the harmonics of the valve assembly or spring if the progressively wound spring does not change spring rate as a progressively wound spring normally is expected.

    Your other concern that there will be 'binding' of the spring coils that touch doesn't ring true to me. I design reciprocating cryogenic machinery including valves similar to engine valves. These machines are dry running so they can't take advantage of the lubricity of the oil to prevent damage from the coils rubbing against one another. The springs I design are always the conventionally wound type and of course, they rub at each end and can wear slightly at those points, but there is no lubrication such as engine oil. We can do a couple of things to help reduce that rubbing but in the end, it isn't worth it. The springs are not significantly damaged by this dry rubbing.
    Last edited: Nov 25, 2011
  12. Nov 26, 2011 #11

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    I am living proof ya can teach an old dog new tricks..I even learned to put the toilet seat down one time..
    I thank you for the excellent post and when back to my notes
    Progressive springs regarding suspension do in fact coil bind as intended.
    I should have refered to the above spring as variable rate since the spring rate goes up the more it is compressed. The following information is from the guys that should know and I hope it clears up some of my cloudy points,,

    What You Need To Know About Valvesprings - All About Valvesprings - How To Do It
    There's A Lot More To Valvesprings Than Seat And Open Pressures. Here's What You Need To Know.
    From the November, 2010 issue of Hot Rod Magazine
    By Marlan Davis
    Photography by Comp Cams, Marlan Davis

    Read more: http://www.hotrod.com/techarticles/...about_valvesprings/viewall.html#ixzz1eoSjVG2c


    Calculating Spring Load Changes
    What if you have to deviate from the catalog's published installed and open pressures? How do you recalculate spring loads if you want or need to run different spring heights? The math is easy for conventional constant-rate (same top and bottom diameter) springs. Let's walk through it: Suppose a particular spring's published specs are: 100 psi closed at a 2.0-inch installed height and 400 psi open at 1.0 inch. What would its closed and open pressures be if it were installed at 1.8 inches and its open pressure checkpoint was now 0.800 inch? This is where spring rate (in lb/in) comes into play. To determine spring rate absent published specs, subtract the original closed pressure from the open pressure, then divide the result by the distance between the original installed and open heights. In our previous example: [(400 - 100) ÷ (2 -1) = 300 lb/in.], multiplying the rate by the change in height yields the change in pressure: [300 × 0.2 = 60 psi]. In this case, since the heights are shorter, you would add the change to the published figures to establish the new pressures (in our example, 160 open/460 closed). If the new height is longer, subtract the change from the published figures. If the closed and open heights did not change by the same amount, solve each case separately. Getting a little more pressure out of your springs could delay valve float, as long as by reducing your heights you don't stray into coil bind.

    Coil Bind
    In any spring installation, coil bind is to be avoided at all costs. If the coils stack solid or bind at or before full lift, at a minimum, the now-infinite load on the valvetrain will cause its weakest link to fail. If you're lucky, the result is merely a bent pushrod. If you're not, you're looking at a broken spring, a dropped valve, or worse.

    When installed at the correct height to develop the right seat and open pressures for the application, the spring needs to have a safety margin before coil bind occurs. The simple formula used to determine whether a spring has sufficient coil bind clearance is: [valve-spring installed height on the seat - (cam lobe lift × rocker arm ratio) + valve lash - safety margin] gives you the remaining open spring length, which should be equal to or greater than the spring manufacturer's published coil bind height.

    Thinking has evolved on how much safety margin is needed. About 0.060 inch used to be the textbook minimum, with more OK and even desirable. That's still an acceptable standard for everyday performance use, but Massingill says that in some cases "0.060 has become the maximum rather than the minimum." Godbold notes that "from high-speed video and testing, it is clear that adjacent coils contact as you approach the valvetrain limiting speed. Hence, modern springs are designed to run near coil bind and use the coil-to-coil interaction for improved damping at or near max lift. This interaction is one of the most effective means of dampening spring surge, but the valvespring must be properly designed in terms of solid stress to safely use this interaction." Depending on the intended use, the spring and cam-lobe design, and the engine builder's preferences, you will now see coil bind safety margins vary from as low as 0.015 inch to as high as 0.120 inch, with tighter numbers predominating on very stiff valvetrains. In a serious valvetrain, anything more than 0.150 inch can cause spring surge, which can greatly reduce the available spring load needed to close the valve.

    Q, thanks for keeping me honest!!
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook