Predicting valve float

[blackjack]

Hey guys cause I don't have my dynamics books handy, just wondering how one would go about predicting valve float for an OHV and OHC engine? The reason is that we want to play around with spring stiffness to maximize power.

Cheers
Steve

[blackjack]

Hey guys cause I don't have my dynamics books handy, just wondering how one would go about predicting valve float for an OHV and OHC engine? The reason is that we want to play around with spring stiffness to maximize power.

Cheers
Steve

[John Stimpson]

If you can get the valve spring specifications from the manufacturer, they'll tell you the natural frequencies, and as-tested ranges where surge comes on and goes away.

Theres much, much more to it. A Spin-tron or similar would really be sweet.

Its GOOD you're considering these things, as spring surge is super bad news. I migh add, however, that if you're using one of the bike engines, with a valvetrain even fairly similar to stock, and you use stock springs, you'll be good to go. Honda does their homework!

[John Stimpson]

Also, valvesprings made of OTEVA-70 or OTEVA-70SC are the best - at least IMHO.

[Dan G]

They had a video in one of the booths at the PRI show this year of a head with a viewing window cut in to see the valve springs in super slow-mo. It was crazy to watch the valve, spring, and retainer all rotated at different frequencies, and to see the wave motion in the spring. I think they posted them to their website but I don't remember the company. Really cool to actually see this stuff happen, some of it is hard to visualize or believe it really is behaving in that manner.

EDIT: found the link...

http://www.racingsprings.com/movies.htm

Check those out.

[Wesley]

From an RPM/force aspect, just get your cam ramp angles, figure out your follower/valve accelerations, and figure out when the force due to the imposed valve motion overcomes the spring force.

Analyzing the harmonics would take a ton of effort, and for not much gain (as compared to testing) especially since we're not spinning any of these motors that fast (or in most cases, even beyond their design intent)

[VFR750R]

Who says valve float is bad?...although bounce usually is.

[blackjack]

Thanks heaps for you input guys. Very much appreciated http://fsae.com/groupee_common/emoticons/icon_smile.gif

Well it's not quite for a FSAE engine, but a L67 (buick supercharged 6). You see the problem is that when updating to higher ratio rockers (lighter after market type) you run into troubles of valve float, depending on the condition of the springs, whereas if you use say modded factory rockers this isn't the case. Just wondering if there was an "easy" formula out there that I could use to predict this, such that I can work out the maximum ratio in which I can increase the ratio by. Please bare in mind that I don't want to change to stiffer springs because of cost and the fact that they hurt HP

On another note I have been active in FSAE-A in the past, was on the 2003 RMIT team to be exact. http://fsae.com/groupee_common/emoticons/icon_smile.gif Cam up with a couple of innovative ideas which were looked down upon that year, but was funny that some of them were used in the following years...

If i was building up a FSAE engine would have no hesitation to boost it http://fsae.com/groupee_common/emoticons/icon_smile.gif No not with a normal turbo or supercharger, but an industrial screw compressor running +60psi on a small diesel engine converted to run on petrol. Restrictor? What restrictor LOL! However in the event the restrictor was an issue build it out of copper and just keep it super cool...

In saying all of this though it probably wouldn't be the most fuel efficient car on the track http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

Been a while since I last read the rules so I could be waaaay off. None the less I have to build a road engine like this one day http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Wesley]

If it were me, I'd get on a Buick forum and say,

"What cam profiles and rocker ratios can I get away with on XX Buick V-6"

Much easier than running numbers, and more accurate.

The good thing about hot rodding with production vehicles is that you can ask any Jimbo down the street what he's running, and if it works, you don't have to do any calculations, even if he's a High School dropout. http://fsae.com/groupee_common/emoticons/icon_smile.gif

[VFR750R]

The problem is quite complex that you ask, considering rocker ratios(and motion ratios), cam profile, pushrod weight, rocker inertia, and stiffnesses of all parts must all be factored in. Considering two aftermarket rockers of the same 'listed' ratio can run measurably different I'd say it's next to impossible without a valvetrain simulation software package like that of Lotus, Ricardo, or GT power. Then you must measure weights and stiffnesses of all the parts in question.

Consider this, drag racers without use of spintrons and valvetrain software run up to 350lbs of seat pressure and 1200+ over the nose. Nascar engines turning close to the same rpm have ~150 on the seat and less then 600 over the nose, but we have access to those things.

You'd probably be better of getting a new cam from comp telling them what rockers you're going to run and stock or aftermarket springs. Comp has simulation software. It'd make more of a difference then rockers alone anyhow.

[blackjack]

"The good thing about hot rodding with production vehicles is that you can ask any Jimbo down the street what he's running, and if it works, you don't have to do any calculations, even if he's a High School dropout"

I work in the aftermarket performance industry, and don't personally like going off "guestimations" I'd much rather talk to fellow engineers about engineering problems. In saying this you'll be surprised what some "high school dropouts" actually know. If a person has a strong enough desire they can pretty much master anything http://fsae.com/groupee_common/emoticons/icon_smile.gif And I'm always there to listen to them http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

Higher ratio rockers are an excellent performance upgrade because there's much less work involved given the amount of gains. Granted never good as a cam change, but not everyone has the competence or the $$$ to undertake such a task. Hence why higher ratio rockers are a good thing for people with little skills or $$$

Once again thanks heaps for the feedback you guys are ace http://fsae.com/groupee_common/emoticons/icon_biggrin.gif I really miss my FSAE days http://fsae.com/groupee_common/emoticons/icon_frown.gif

[Mike Flitcraft]

If jack's working on a longitudinal/fwd Buick, I can't blame him one bit for wanting to do the rockers and not the cam.

If he's working on a RWD....why the hell haven't you opened up the block yet?

[Kurt Bilinski]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Dan G:
EDIT: found the link...
http://www.racingsprings.com/movies.htm

</div></BLOCKQUOTE>

bad link

[Wesley]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by blackjack:
In saying this you'll be surprised what some "high school dropouts" actually know. If a person has a strong enough desire they can pretty much master anything http://fsae.com/groupee_common/emoticons/icon_smile.gif And I'm always there to listen to them
</div></BLOCKQUOTE>

That was sort of my point.

And if you don't like going off of guestimations, fine, but theres a lot of good test data - in the form of real world performance - that justifies modifications way better than any computer simulation.

[blackjack]

Clear, credible test data is always a great thing as somethings are extremely difficult to model accurately. I have a couple of friends who with CAE for living, one of has spend many many weeks trying to optimize the spot weld properties for the new Camaero during crash analysis. I suppose I'll have to figure out whether or not I can justify calculating this theoretically.

Cheers
Steve