2/7 allready incorporates some overlap.
Maybe not enough.
Less overlap = lower float rpm.
Less overlap = slower valve up speed possible.

The float rpm I "calculated" was pretty high.
This suggests 2/7 could be too small, try 3/8?
Or your 2/5?
What are the float rpm's with those numbers?
(now you do the math )


Still, there is - must be? - an rpm where the cams spin fast enough to be too fast for the valve spring moving up but too slow for the valve to ride the top of the lobes, IOW not close at all.
It's not that valves are fast enough and suddenly, like pressing a button, they are slow.
Springs do not work like that.

And in this timeframe they hit the seat uncontrolled.
And that's bad.
For retainers.

 

2/7, 3/8, 2/5, π[sup]-1[/sup], whatever. Valve duration doesn't to tell you anything about float.

As far as the must be range, I'll guess that it's from 15,000 rpm to 40,000 rpm, +/-5000 rpm. I doubt the valve will ride the tops of the lobes ('surf') at less than 35,000 rpm.

No comment. (This isn't a snarky statement. I simply have no opinion about this statement.)

I disagree. It's possible that the valve float a little under design conditions. It's covered in the valve loft link I posted before.

 

so back to some down to earth questions:

1. does valve float occurs in the normal RPM range? (less than 9,000)
If YES - is it bad? dangerours to the engine?

2. do stiff valve springs reduce/prevent valve float or they are good just for over rev protection?

3. what spring rate is reqired for the above?
i see Supertech offer few options but BC and Eibach offer just one option of ~84lb springs. Somewhere else Alex says that this is not enough to protect from the over rev - would you agree with this claim?

 

Regarding points 2 & 3, stiffer springs do prevent valve float which as a result provides over rev protection. The stiffer the springs the higher the threshold for valve float, but the spring pressure also causes wear. So there's a trade off between the two.

 

Valve actuation and valve operation reminds of the "Whack-A-Mole" game at the fair grounds. If you always knew which mole is going to pop its head up, you keep pounding on that one. You can then do it faster and faster and whack him before his head gets all the way to the top. Do it consistently and whack him before he comes up fully and you "float" his little head at mid-stroke.

 

The point of stiffer springs is to make the valve follow the cam at higher (= above official redline) rpm's so the cam closes the valve controlled and it does not hit the seat at speed.
This is bad for retainers and seats.
Valves do not float, as in "ride the top of the lobe", at rpm's the engine can reach without disintegrating.

Take an absolute safe rpm, let's say 8800 rpm.
At this point we are sure all valves are closed when they have to be as the ECU still injects fuel and fires the sparkplug, and the engine produces power.
Not peak power but it produces power.

Following dwight - and why not? - we'll use 2/5th cam rotation to open and close the valve and 3/5th cam rotation without valve action.
This means 1/5th is closing time, unless the 2/5th is not devided equally, but let's assume it is for now.

8800 crank = 4400 cam rpm = 73 cam revs per second (don't mention the 0.3 difference.. please)
So 1 cam rev = 0,0136 sec.
1/5th of this = 0,00273 sec.

The same for 9000 crank = 4500 cam = 75 cam revs per second.
1 cam rev = 0,0133 sec.
1/5th of this = 0.00266 sec.
Let's call this valve top speed.

Now, what happens at 10700 crank = 5350 cam rpm?
This is 89,166 rev per second or 0,0112 second per rev.
To ride the top of the lobes, the valve has to still be closing during this time.
Yet, we have seen it is able to close 4, YES 4!!! (actually 4-and-a-bit) times faster than this at 8800 rpm.
0,0112 / 0,00273 = 4,102.

So at this 5350 rpm the valve spring just thinks..... ohhh F**K it, I'm done, they can all go to H**L
I'll close when I feel like closing.

I
DO
NOT
THINK
SO.

 

How do valves get bent during an over-rev? I'm sure there are s2000's that have bent valves at much less than the RPMs you're saying are required for valves to float?

You obviously have a much better grasp of this concept than I do, but I'm still left with questions.

Do valves hitting their seats "at speed" really cause damage? Is there proof besides the cracked retainers you're claiming result from the unchecked closing of a slightly "floating" valve.
Does the valve follow the "back" (closing) side of the lobe at 9000 RPM (or any other given RPM)?

And I again, I ask, because it hasn't been answered yet. If it takes 10,000+ cam RPM to truly float valves, how are valves bent during an over-rev?

If what you're suggesting occurs, where the valve closing falls slightly behind the cam profile, wouldn't this cause the valve (through the rocker/cam follower) to meet with the cam profile at a different point in each rotation (of the cam)? Eventually, couldn't this reach a point that the valve is not closing all the way?

 

so what you say?

1. NO ???
2. just for over rev protection ???

 

From my understanding, he is saying that valve float of any kind does not occur at or below 9000 RPM (with the stock valvetrain) and stiff valve springs do a better job of keeping the valves following the cam profile at higher (than stock) RPM; this is equivalent to protection from some kinds of over-rev damage. Stiffer valve springs do nothing to protect the bottom end, and those parts may still be subject to damage during an over-rev.

 

I find this post very hard to read.

First, I'm a stupid American and I only recognize a period (.), also known as a full stop on your side of the pond, as an acceptable symbol to represent radix points.

Second, I don't see any need to convert anything into periods (cycle time, not the period/full stop previously mentioned) (i.e. 0.0126 sec). Everything can be said in terms of engine rpm and people are familiar with that. In addition, I don't like cam rpm because it seems to confuse some people. Depending on the calculations involved, you could make a point it would be better to talk about this all in terms of cam angular velocity with the unit of radians*sec[sup]-1[/sup]. Physics and engineering-wise that makes but I'm sure that it would be harder for people to follow.

Third, and stiffer springs don't necessarily perform better at higher speeds. I've mentioned this twice before, but it's possible to utilize valve loft to have the valves float open a little. This can increase lift at high RPMs which thereby increase flow. If the springs were stiffer, the valves will just follow the cams and not have the increased lift at higher engine speeds. Either you'll need to do have a good understanding of calculus to understand why this is safe or you'll need to do all of the calculations alluded to you in post #66 and to which you replied:


Finally, I'm don't know who here is claiming that the valves can surf the cam lobes. I don't know why you're trying to dispel that 'myth'. You might as well be telling people that past 10,000 rpm the cams don't turn into werewolves and eat the retainers so getting silver retainers won't fix the problem. I've already stated that surfing would take about 40,000 rpm. That should be painfully obvious because it's simply ~9,000 rpm X 4 strokes ≈ 40,000 rpm which is significantly higher than 10,700 (engine) rpm condition which Billman250 measured. You don't need to do all kinds of strange conversions. That's unnecessary and it just serves to make your post harder to read.

In summary:

• I disagree regarding valve float. It isn't always bad.
• I agree that it's not surfing at 10,700 rpm. I'm curious as who doesn't agree.