Ok, maybe a good understanding of calculus isn't needed to understand why valve loft (controlled valve float) is ok.

Valve loft is like jumping a table top at the proper speed (ignore vehicle pitch). As long as you land on the decline, you're fine.
Note: Cam lobes aren't table tops; they are smooth. You don't need to decide between not jumping them and jumping them. These is no medium speed where you jump and come up short and nose in to the table top.

 

Ok so Dwight is it safe to assume that the F20c camshafts do not caust valve loft? After reading and understaning what it is, i think its an extreme application used in high end motorsports. That being said its not relavent to the F20C.

Being an open minded person im trying to consider all three sides.
Billman250- At a certain rpm the valves do NOT close all the way, concluded from his own tests
Dwight- There are certain circumstances that cause valves to not follow the cam profile and "float" "loft"
SpitfireS- Mathematically there is just not enough time where the cam is out of cotact with the valve for it to float at all


All of those points being taken into consideration, as well as metal harmonics and the bouncing/spinning i saw from the videos posted above leads me to think its a number of things that add up to cracked retainers/bent valves and "float"

I think at certain rpms the valve is not staying closed long enough to seat 100% and causes it to wobble/spin

I also think at certain rpms due to the harmonics of the engine as a whole, the valve is slightly bouncing on its seat before finally sealing. Causing more spin/wobble

Lastly i feel in an over-rev situation there is an rpm that causes all these things to happen at once and the valve never touches its seat. Its like if you dropped a straw down a hole at an angle and it bounces off the sides the whole way down, causing all kinds of wear and tear. Not to mention it takes longer for it to reach the bottom.

I don't think the cam is hammering the valve open so hard and fast that it just hangs out there and gets hit again before it can follow the profile of the cam. I think due to vibration/harmonics/wobbling it is not able to close at the correct speed and thats when the damage happens.

Again no math/physics to back this up. Just emperical evidince from above posters and the videos that i have seen.

 

That's not quite right. Given that we're talking AP1, and by overrev I mean about 10,700 rpm.
My position is:

• As stated by Billman250, AP1 overrev causes retainers to crack and replacing with AP2 ones seems to prevent this. This is important for people to know and the exact cause is secondary because we already have a solution.
• Under normal operation the S2000 probably doesn't use valve loft, but it's possible that it does so the valves don't always follow the cams. If it does use valve loft then stiffer springs prevent this and you lose performance at higher engine speeds.
• There is no way there is no way a regular overrev is enough to have the valves surf/skip across the tops of the valve loves. This would require more like 40,000 rpm and you'd bend the valves due to interference way before then (guessing 12,000-15,000 rpm). Any further discussion regarding this is a waste of time.
• Billman250's test showing the valves not closing at overrev may be flawed. As I understand he only used a head. As the valve closes at the end of the intake stroke and during the beginning of the compression stroke, cylinder pressure increases so there's a force applied to the bottom to the valve which may aid in closing it.
• At 10,700 rpm, there may be a vibration/resonance issue, but there is insufficient data. If there is a vibration problem, stiffer springs may provide more overrev protection by moving the natural frequency higher. However, excessively stiff springs may cause wear problems.
• Retainers may be cracking due to an phenomenon not mentioned.

 

Every rpm above redline is an overrev.

An overrev between redline and piston & valve contact - the exact rpm range is not clear - will damage AP1 retainers.
Why?
Due to uncontrolled landing of the valve, it will hit the seat at speed.
The valve can bounce and wobble afterwards, like 4forall mentioned, the valve spring can vibrate, dance and rotate, I've seen the video's.
They are nice to watch.
Valve loft may happen too as the cams are spinning faster than designed so the valve may open a bit deeper than normal, this will increase the closing distance - and end speed! - but the valve assembly has time to close during the dwell period, the base circle cam time.
I think I've shown this by looking at the pure timing in seconds, or milliseconds.
It will not close controlled by the cam profile anymore though.
Hammering of valves that not close by the cam coming around again is not possible at the rpm's were talking about here.
There is a good chance there will be valve seat damage.
So, a valve train with cracked retainers may not be serviced fully by just replacing retainers.

An overrev that will cause piston & valve contact will also crack AP1 retainers, the piston & valve contact is way worse so one has to swap the head anyway.

Replacing AP1 retainers for AP2 retainers gives extra protection against cracked retainers during an overrev.
This only helps if the overrev stays below piston & valve contact rpm.
It will not make the valve close any quicker though.
So hitting the seat at speed, valve loft, bouncing, wobbling, rotating and dancing can still happen during an overrev.
Seat damage after multiple overrevs - because with AP2 retainers it's now safe to overrev.. right? - is very possible.

I may have used other words, but I've been saying most of this since my post #1.

Btw, in the article where dwight found the cam profile picture - circletrack website - and where valve loft is described there is not a single word about hammering valves.
Or cracked retainers.

 

Do you rev your engine to 9000 rpm? It that overreving your engine? F20C redline is 8800 rpm. By your definition, 8850 rpm is overrev.

 

Every rpm above the OEM ECU fuel cut off rpm is by defenition an overrev.
Do I rev to 9k?
All day!


Use an AP1 ECU with an F22 and rev to ful cut-off.
Are you overreving?
From an F22 valve train point of view: not really.
From F22 piston point of view: most likely.

IIRC the OEM F20 ECU fuel cut-off is at a true 8900 rpm.
(could also be 8897 or 8905 or 8896 or 8909 or... do I have to continue?)
That is what Modifry mentioned in a post and I trust what he posts.
The dash is a little enthusiastic in displaying rpm so many will say the F20 revs to 9k.
Don't try to get me on 100 rpm, or 200.

 

You didn't answer my question. Maybe it wasn't clear enough.

 

I will try again.

First the Red part:
After searching for +30 minutes I failed to find this post where Modifry states the OEM AP1 ECU cuts fuel at 8900 rpm.
He also states the dash tach reads high.
I know I've seen it so I'm going to stick with it unless someome proves otherwise.
Where does your info come from?
Digging through a lot of posts I found statements about overshooting fuel cut-off rpm as well.
It depends on the gear you're in and throttle position.
Assuming Honda knew and tested this one could think an overrev to 9400 true rpm is safe.
It is still above fuel cut-off so still an overrev.

The Green part:
I stand by my definition that every rpm you can not reach accelerating after fuel cut-off is an overrev.
If I use my redline (8900) 8850 is not an overrev.
Using your redline it is.
If you take the overshoot into account: you're OEM overreving.

The Blue part:
Maybe during an overshoot bouncing off the fuel cut-off in 4th?

The Black part:
If I reached a true 9000 rpm during a mechanical overrev or overshoot while bouncing off the fuel cut-off, then yes, 9000 is an overrev.
Using a true 8900 as OEM fuel cut-off.

Btw.. I see 2 question marks in your post

 

Clearly, your "All day!" was the answer to the "Do you rev to 9000 rpm?"
As a result, when I said you didn't answer my question, I thought it was obvious that I was only referring to whether or not that was an overrev. I apologize for assuming that you would have been able to make that distinction.

In the future any question I pose to you shall be color coded.

 

I think I missed it, didn't I