valve train springs 1

[fatman57]

What scope is there to making custom valve springs that retain the factory Tensile strength but lower the energy needed to push the spring down? Is this a bit of an oxymoron?

 

[ivymike]

to keep the tensile strength, use the same material (and same wire diameter and bend to the same radius).

to reduce the energy required to lift the valve, any combination of the following:
1 - reduce max lift
2 - reduce fitted force of the spring
3 - reduce spring stiffness

1 will reduce engine performance.
2 might not be possible with the geometry and fitted force you have, may reduce your engine speed capability, and may increase the working stress range of the spring to an unacceptable level
3 will reduce the surge frequency of the spring and the speed capability of the engine

 

[ivymike]

strike the bit about working stress range in (2).

 

[fatman57]

thanks.

what about a change of materials?

is spring stiffness reduction achieved by having less windings?

if i used the same material but a smaller wire diameter would i have less tensile strength but also need lees energy to push the spring down?

 

[BrianPetersen]

Spring rate is affected by:
- Diameter of the wire (smaller = lower spring rate)
- Number of active coils (more = lower spring rate)
- Wound diameter of the spring (bigger = lower spring rate)
- Material properties (but you need something very similar in properties to spring steel and all steels have approximately the same elastic properties which is what affects the spring rate, only the fatigue and yield properties vary, but spring steels are chosen as such for a reason, so for realistic and normal applications there is not anything productive that can be done with this)

You need a certain amount of "seat" force on the spring in order to ensure that the valve remains closed and doesn't bounce off the seat. You need a certain amount of force on the spring as it goes over the nose of the cam lobe at maximum possible engine RPM plus a good factor more, to protect against "valve float" in which the follower leaves contact with the cam lobe. Those two forces at known compression values of the spring will define the spring's necessary spring rate and free length. When you start getting subject to spring surge and the like, it gets really, really complicated. Very often, valve springs are chosen so that at maximum lift, the coils are not all that many millimeters away from "coil bind". Reason is that the spring can't bounce around much if it's almost completely contained.

Don't neglect the effect of the cam lobe profile. A profile that holds the valve near max opening for quite a while and limits acceleration over the top of the lobe won't need as much spring force to overcome valve float. But, unless a really long duration is acceptable, this forces large accelerations off the seat and approaching the seat. The geometry of the lifters may dictate some limitations in this regard. (you can't have concave cam profiles with flat-tappet lifters, for example, and you can't have the cam profile push too close to the edge of the lifter without really bad things happening.)

It will probably help if you explain exactly what it is that you are trying to accomplish and why, and exactly what the application is.

 

[ivymike]

don't forget that the springs don't waste all of the energy put into them- you get nearly all of it back on the closing side of the cam lobe. The big losses are generally from concentric springs rubbing against each other and from friction between valvetrain components (which is reduced by lower spring force).

 

[evelrod]

Camshaft lobe lift profile plays a HUGE part in what type of valve springs applicable. If one were to really get serious in redesigning the lobe lift/rate to a "softer" profile it just might be possible to make it possible to reduce the spring rates and NOT loose performance. I said it MIGHT be possible. A big question is why would you want too unless it was for the far out fringe of performance.
In the 80's I did my own (with a lot of help) cam design for out Lotus twincam race engines. The purpose was to increase overall performance while making it possible to retain the 85lb seat pressure and 250lb. over the nose. All this is academic these days, but we actually accomplished the task and those profiles would be adequate for today in vintage racing.

Bear in mind that what Isaac says it truth...Do you really want to reduce performance just to get a lighter spring rate?

Postivymike (Mechanical)
17 Dec 09 13:16
don't forget that the springs don't waste all of the energy put into them- you get nearly all of it back on the closing side of the cam lobe. The big losses are generally from concentric springs rubbing against each other and from friction between valvetrain components (which is reduced by lower spring force).

Rod

 

[fatman57]

thanks BrianPetersen! much obliged!

i'll be honest and say i was thinking of efficiency & economy. i'm servicing a cylinder head and doing some further research. Ideally I would like to be able to crank it with my hand but there are many things in my way!

it is ture that alot is given back, and in the pinch of it all the frictional losses might/can be avoided.

i've seen springless systems and there are many different systems in the past that have been made. Alot of HP can be saved via the valvetrain plus things like a smaller battery etc as it won't take as much to crank.

 

[fatman57]

i guess i'm looking more at the road car than performance racing! but its all related really!

i thought it might be possible too but thought i'd see what others thought or had experienced.

changing/altering the springs is a bit easier then redesigning the valve train.

 

[evelrod]

You answered your question in your first post.
Anything further is starting to look quixotic.

Rod

 

[desertfox]

Hi fatman57

I agree with the others that its not as simple as you think,firstly the spring is designed to operate at a relatively high temperature so any material changes would have to take account of that.
Most valve springs are carbon steel and I doubt you would reduce spring stiffness significantly by replacing the spring based on material alone unless you used a non-steel.
Then there are the manufacturing tolerances to consider on the physical size of the springs and there affect on the actual spring force/stiffness values across a set of springs.
Stresses,Fatigue will all need re-looking at as previously mentioned but the other major problem I see is surge and spring resonance.
If your spring happens to have a natural frequency close to the operating frequency,then it can cause bounce which can create a lower spring force than predicted at minimum spring deflection or alternatively deflections of individual coils are greatly increased along with the torsional stress in the wire.
To avoid resonance they suggest the spring should have a minimum natural frequency 13 times greater than the operating frequency.
Final nail in the coffin is surge,a surge wave is estabilised which transmits torsional stress from the point of loading along the spring to a point of restraint.
The surge wave limits the rate at which the spring absorbs and releases energy thereby determining whether the spring stays in contact via the valve rod with the cam.

desertfox

 

[ivymike]

The surge and resonance you're describing are the same phenomenon.

 

[patprimmer]

Frictional losses in the valve train are insignificant when compared to losses between the rings and the bores.

A roller cam significantly reduces valve train losses.

Lighter components reduce valve train losses.

These tend to greatly increase costs or reduce life or both.

Modern multi valve designs have some advantage in reduced weight and required lift but increase costs.

OHC also reduces reciprocating weight in the valve train.

What do you think might be the total reciprocating weight in the valve train vs total reciprocating weight in the piston rod assembly. Also what distances are these components moving relative to each other.

What losses will you get if the valves bounce.

What losses will you get if you need a significantly larger engine to make the specified power.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[desertfox]

Hi ivymike

Yes my mistake, thanks for pointing it out.

desertfox

 

[SomptingGuy]

What do you think might be the total reciprocating weight in the valve train vs total reciprocating weight in the piston rod assembly

It's possible to balance the reciprocating masses of the piston/rod assembly given enough of them and/or some balancer shafts. But it's not really possible to do the same for the valves. Not related to losses I know (NVH really), just being pedantic.

- Steve

 

[patprimmer]

I'm referring to losses due to inertia with the constant acceleration and deceleration as the parts change direction of travel.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[SomptingGuy]

Uh?

The acceleration and deceleration will put heat through the bearings I guess. But otherwise the energy can't really go anywhere else.

- Steve

 

[dicer]

It is very difficult to find the perfect spring.
In operation they are "ringing" all the time. Given the correct cam profile and speed, components could be overloaded. Its not an exact science.

 

[patprimmer]

Valve springs get reasonably hot and are normally cooled by engine oil.

My real point meant to be that the losses are small compared to pistons and rods and rings.



Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[dgallup]

quote

Modern multi valve designs have some advantage in reduced weight and required lift but increase costs.

Quite a few years ago Honda claimed that they could produce a DOHC 4 valve head for the same cost as a 2 valve head. Smaller valves, smaller springs, optimized materials and production processes.