[Craig Hendrickson]

I noticed another thread with a question about cylinder head flow and how computing the "Area Under Curve" (AUC) of the head flow curve might help evaluate its potential performance. There are several AUC calculators available on the web, notably, John Wallace's which was linked in the original thread. It appears that all or most of these use PHP or somesuch backend server language to do the actual calculation. This of course requires an internet connection and an operating back end PHP (or whatever) server.

The calculator I just wrote and posted on my website is written in Javascript. So, all you need is a decent browser with Javascript enabled (attention Firefox "NoScript" users!). You can also download that single page to your computer and run it offline. Since I put it out as free software, feel free to modify it for your purposes, i.e., adding more or finer data points.

My purpose was not to "reinvent the wheel", but to have this software be the first version of a program which will, in a later version, matrix the lift/flow data with cam degree/lift data. A following version could matrix the flow/cam data with piston size/position data, hopefully yielding a "flow potential" value of the whole combination. These programs are NOT intended to replace computer programs simulating running engines.

Here's the link to the AUC calculator. Feedback is welcome.

http://www.originalho.com/AUC4HFcalculator.html

[johnta1]

Cool calc, Craig!



I could never get using javascript, easier for PHP to me.

[Tom Vaught]

Understandable as you do not have the computer science background in the industry that Craig has. (That was his Third Job

1) Air Force Officer

2) HO Racing

3) Computer Software Expert

Tom Vaught

[Craig Hendrickson]

Originally Posted by Tom Vaught

Understandable as you do not have the computer science background in the industry that Craig has. (That was his Third Job

1) Air Force Officer

2) HO Racing

3) Computer Software Expert

Tom Vaught

LOL!

Actually, I was doing computer programming at USAFA in 1965. There was no Computer Science major at that time, but it was requirement for my Astronautical Engineering BS degree. The only precise way to calculate orbits and optimized rocket stages was with a numerical digital computer -- analog equations and slide rules were not accurate enough, even though common at the time. I'm retired since 2008 and I've pretty much seen it all over the years. Don't even get me started one some of the cr@p (languages, applications and operating systems) that are out there now.

Sorry for the thread drift.

The kind of feedback I am looking for on this little application is what are real racers looking for as a tool in this area? I'm willing to do some additional programming as freeware in HTML/Javascript (even though I have expertise in many other areas) since that is what is used in a standalone browser program today.

[Craig Hendrickson]

To give you guys an idea of where I'm thinking of going with this application, let's take the next version which deals with cam degree and valve opening.

Unfortunately, few of us have access to the actual lift curve of any camshaft lobe, so it is even more indeterminate than the head flow data. But, some vendors like Comp Cams provide a catalog with the following lobe profile information:

lift@0.006in in degrees duration
lift@0.050in in degrees duration
lift@0.200in in degrees duration
lift@max (whatever that is) at whatever centerline is specified

This is four data points and can be "curve fitted" with the following equation, plus allowing for "end conditions" like zero lift before 0.006in (tappet accleration has been zero) and nose lift (where tappet acceleration is zero and reverses):

y = ax^3 + bx^2 + cx + d ... plus end conditions

Once an approximate curve is fitted, this means that any lift point on the lobe or valve (accounting for rocker ratio -- theoretical or real) can be calculated.

JavaScript can do this math (with proper programming). So, this means that the valve open position at any crankshaft position can be calculated and multiplied in some fashion by the air flow at that valve lift.

This should be food enough for thought by real racers.

[johnta1]

this means that the valve open position at any crankshaft position can be calculated and multiplied in some fashion by the air flow at that valve lift.

Add piston speed also?

[Craig Hendrickson]

Originally Posted by johnta1

Add piston speed also?

Sure, why not? You of all people know that I have the equations for that from the TD-02 book.

[68gtoMN]

Still a wannabe racer yet impressed nonetheless. Craig, your hobby now will expand the knowledge for others and keep this alive for us who are not "there" yet.

Thank you and Kern...Mr. Wallace also

[Craig Hendrickson]

Originally Posted by 68gtoMN

Still a wannabe racer yet impressed nonetheless. Craig, your hobby now will expand the knowledge for others and keep this alive for us who are not "there" yet.
Thank you and Kern...Mr. Wallace also

I'm more than pleased to provide software for real racers or maybe even wannabe racers who might someday be real racers. Since not everyone can be a computer programmer, I'm happy to try and fill in the void.

[Craig Hendrickson]

Originally Posted by OriginalHO

(snip)
This is four data points and can be "curve fitted" with the following equation, plus allowing for "end conditions" like zero lift before 0.006in (tappet accleration has been zero) and nose lift (where tappet acceleration is zero and reverses):

y = ax^3 + bx^2 + cx + d ... plus end conditions

(snip)

A couple of corrections I caught in the "end conditions" in my previous post:

"zero lift before 0.006in (tappet acceleration and velocity has been zero)"

"nose lift (where tappet velocity is zero and reverses direction)"

I'm surprised nobody caught this. The equation must have glazed over everyone's eyes. :-)

[Craig Hendrickson]

I made an upgrade to the AUC calculator. I added a "select box", which allows one to pick a typical Pontiac head and the flow data will be propagated to the flow data boxes. The flow data is from Stan Weiss' flow charts available on the web.

If you download the html page to your computer, you can edit the values in the select box. See line 71 and following in the source code. With an ASCII editor (Notepad, Wordpad, Textpad, etc), just change the comma-separated values and name of the head to whatever. You can also add line(s) too; just use the existing values as a guide.

Please be aware that with JavaScript, one cannot read or write to disc because that was deemed a BIG security risk. Well, actually HTML5 allows this, but this just came out and most browsers do not support it. Editing the SELECT values is really no more trouble than having a separate editable file with the flow data that is read by the application.

[johnta1]

Cool addition, Craig.



I sent you a PM.

[65nss4spdGTO]

Originally Posted by OriginalHO

To give you guys an idea of where I'm thinking of going with this application, let's take the next version which deals with cam degree and valve opening.

Unfortunately, few of us have access to the actual lift curve of any camshaft lobe, so it is even more indeterminate than the head flow data. But, some vendors like Comp Cams provide a catalog with the following lobe profile information:

lift@0.006in in degrees duration
lift@0.050in in degrees duration
lift@0.200in in degrees duration
lift@max (whatever that is) at whatever centerline is specified

This is four data points and can be "curve fitted" with the following equation, plus allowing for "end conditions" like zero lift before 0.006in (tappet accleration has been zero) and nose lift (where tappet acceleration is zero and reverses):

y = ax^3 + bx^2 + cx + d ... plus end conditions

Once an approximate curve is fitted, this means that any lift point on the lobe or valve (accounting for rocker ratio -- theoretical or real) can be calculated.

JavaScript can do this math (with proper programming). So, this means that the valve open position at any crankshaft position can be calculated and multiplied in some fashion by the air flow at that valve lift.

This should be food enough for thought by real racers.

With the additions of products like Audie Technology, Cam Pro Plus, it gives us much more information with a cost effective system.

Calvin Hill
Hill Performance
708-250-7429

[jimmy daniel]

love reading it but understand none of it! that goodness YOU do.

[Half-Inch Stud]

Originally Posted by OriginalHO

Sure, why not? You of all people know that I have the equations for that from the TD-02 book.

Well, yea, that Velocity vs Crank position had me designing camshafts & exhausts. MY own Excel file showed a NEED to increase OVERLAP to PROMOTE early VELOCITY pull, for early Flow Momentum, since the Slug had NO signal for quite a TDC dwell.

[Half-Inch Stud]

http://www.originalho.com/AUC4HFcalculator.html
This AUC Calc seems to promote Low-Lift Flow: 0.200" lift CFM makes the "biggest Gradient difference" in results (aside from radically increased lift).

[Craig Hendrickson]

Originally Posted by Half-Inch Stud

http://www.originalho.com/AUC4HFcalculator.html
This AUC Calc seems to promote Low-Lift Flow: 0.200" lift CFM makes the "biggest Gradient difference" in results (aside from radically increased lift).

Which cylinder head flow values?

[Craig Hendrickson]

Here's the second program I posted about above. This calculator computes a "best fit" curve approximating the actual lift profile of the specified camshaft lobe. It does so by splicing together three Gaussian normal distribution curves, iterating the standard deviation to achieve "best fit". Using the resultant cam lobe profile, the calculator then computes the duration at each 0.100in increment of valve lift (lobe lift * rocker ratio).

I used the spliced Gaussian curves because it is quicker to calculate than iterating to find the coefficients of the high order polynomical equation. On a "Cam Doctor" printout I have for the HC-03B, it is within one or two degrees of actual.

Here's the link to the calculator:

http://www.originalho.com/CamCurveCalculator.html

Here's some data for you to try from the Comp Cams Lobe Catalog for their Magnum Hydraulic lobes:

Lobe Nbr-Dur@.006-Dur@.050-Dur@.200-Lobe lift
5216-280-230-137-.306
5229-288-237-148-.322
5209-296-246-158-.340

You can also change the rocker ratio to see the effect on duration at lift.

NOTE: This calculator only works for "computer generated" profiles. It will NOT work for "McKellar" designs (9785744) or "NHRA cheater" cams with flat noses (constant lift across some degrees of duration).

[Tom Vaught]

Calculator works good at work, Craig! (;>)

Tom Vaught

[ry57pont]

Craig, you are on the list of people i would like to sit down and have a beer with.