Milwaukee Midget

[Milwaukee Midget]

Midget,

Any response from St. Dema?
 
Fordboy

Seems Buddy might have been bending his ear about the Flatcad last Tuesday on his visit to Dema's shop, which I think was the day my cam arrived!  

Buddy, hope you enjoyed your visit.  I hope to meet him myself someday soon.

News from Harland Sharp is, as of this morning, they do not have the information on their BMC rockers regarding the moment of inertia.  Randy said he's still working on it, and we CAN proceed without that info, but I guess I'd like to know before we move any further on this valve train.

So - turning to the great brain trust that is the land speed racing community - can somebody give me a method of calculating the moment of inertia on a billet rocker?  It's a shaft type, and I'm thinking maybe using a method similar to checking end weights on a connecting rod might be a good starting point.

[wisdonm]

A preachment, dear friends, you are about to receive on John Barleycorn, nicotine, and the temptations of speed.

Don't know much about saints because I'm a Packer fan. But don't forget about these two classics.

http://www.youtube.com/watch?v=GHf7TD4qwjk

[fordboy628]

News from Harland Sharp is, as of this morning, they do not have the information on their BMC rockers regarding the moment of inertia.  Randy said he's still working on it, and we CAN proceed without that info, but I guess I'd like to know before we move any further on this valve train.

So - turning to the great brain trust that is the land speed racing community - can somebody give me a method of calculating the moment of inertia on a billet rocker?  It's a shaft type, and I'm thinking maybe using a method similar to checking end weights on a connecting rod might be a good starting point.

Midget,

Sorry to pop your bubble, but a rod weight checker will not work.  What needs to be measured is rotational moment of inertia.
 
I have plans somewhere for building a simple moment of inertia checker for rocker arms.  Let me see if I can dig them out.   I want to say that they originally came from Dema, but it was so long ago that I am not sure.  I am willing to machine up a moment of inertia fixture/setup to have it available for the future.

I may have some numbers for 'Titan' roller rockers for BMC's.  Will need to mine the 386 (you know the one) for info.

In the meantime, what St. Dema is doing is not dependent on knowing the moment of inertia of the rockers.  What you need from Dema is: the maximum NEGATIVE acceleration for the cam lobe (as the tappet goes 'over the nose'), AND, if there is ANY instantaneous jerk (which would be infinite) designed into the cam profile.  (Instantaneous jerk causes 'spring surge', but, I think that jerk, as a derivative of motion, is somewhat over-rated.)  That info, with the rocker MOI, and the mass (weight) of all the components will allow us to run a comprehensive check of the of the valve spring pressures required to control valve train motion and discern if there is an adequate safety margin.   I'm pretty sure you are OK in terms valve spring pressure & safety margin, but it would not hurt to run the numbers.

Will try to post up some "Cam Porn" once the current home project is concluded.   Before you get too excited, I mean diagrams of cam motion & the derivatives, with some explanations of the theories.
 
Fordboy

[manta22]

MM;

Here is one way to measure the moment of inertia of rocker arms-- connect the rocker arm directly to the shaft of an ironless armature permanent magnet DC motor which is driven by a constant DC current. This gives the motor a constant torque so its rate of angular acceleration will be determined by the motor + rocker arm moment of inertia. Measure the RPM vs time to get angular acceleration. Remove the rocker arm and repeat the experiment to get the moment of inertia of the motor alone and subtract the two numbers.

Regards, Neil  Tucson, AZ

[Milwaukee Midget]

MM;

Here is one way to measure the moment of inertia of rocker arms-- connect the rocker arm directly to the shaft of an ironless armature permanent magnet DC motor which is driven by a constant DC current. This gives the motor a constant torque so its rate of angular acceleration will be determined by the motor + rocker arm moment of inertia. Measure the RPM vs time to get angular acceleration. Remove the rocker arm and repeat the experiment to get the moment of inertia of the motor alone and subtract the two numbers.

Regards, Neil  Tucson, AZ

I LIKE IT!

Looks like a trip to American Science and Surplus is in order.

http://www.sciplus.com/

This might be the ticket?

http://www.johnsonelectric.com/en/products/motion/dc-motors/standard-dc-motors/high-voltage-dc-motors/datasheets/DC971-2-LG-011-metric.pdf

$9.50!?!?  Even if it doesn't work, I'll get hours of joy trying to MAKE it work! 

[manta22]

MM;

The motor you linked to has far too high moment of inertia to get any meaningful data. What you want is a small ironless rotor DC servomotor-- something like the 120-18 series here: http://docs-europe.electrocomponents.com/webdocs/002f/0900766b8002f9c4.pdf 

These are available here: http://uk.rs-online.com/web/p/servo-motors/4329765/ but they cost 52.46 Pounds. You can probably find something like those far cheaper on the surplus market. Don't forget that you need to drive them with a constant current so a 6V to 12V motor would probably be the easiest to drive that way.

If you can find one with a tach output it will make monitoring the motor's RPM easy.

Regards, Neil

[Interested Observer]

MM:
Two methods  of determining the moment of inertia of the rocker come to mind.  The first is to model it in a 3-d Cad or FEA program and let the program calculate the moment.  This would require a reasonably accurate model, which may be somewhat laborious to produce, depending on how complicated the shape is and if there are different materials involved.
The second, and recommended method, is to just measure it!  All you have to do is determine the weight of the rocker, and where the center of mass is (neither of which is all that difficult), then hang the rocker on its side on a bifilar pendulum of a known length, rotate the rocker slightly in the plane of its normal motion and measure the oscillation rate.  From the results, a little calculation gives the moment about the CG, and another calculation corrects it to the center of the rocker shaft location.  Very simple and straightforward.

[Stan Back]

"Very simple and straightforward" . . . but way past my comprehension level.

Go for it, Chris!

[Jack Gifford]

... That info, with the rocker MOI, and the mass (weight) of all the components will allow us to run a comprehensive check of the of the valve spring pressures required to control valve train motion and discern if there is an adequate safety margin...

Well, if you then apply a safety factor of about 2:1 you'll probably be "safe". I went through that whole tedious exercise once; only to scatter valevetrain bits on the track. Various approximations (e.g. spring mass) and frictional forces apparently kill the usefulness of the F=MA calculations. I do, however, now rely on cam grinders who will tell me the maximum negative acceleration of their various profile choices, or will limit that to a specified value for a custom lobe- but I only use that to make comparisons to spring forces of known-reliable engine combinations.

[fordboy628]

Well, if you then apply a safety factor of about 2:1 you'll probably be "safe". I went through that whole tedious exercise once; only to scatter valevetrain bits on the track. Various approximations (e.g. spring mass) and frictional forces apparently kill the usefulness of the F=MA calculations. I do, however, now rely on cam grinders who will tell me the maximum negative acceleration of their various profile choices, or will limit that to a specified value for a custom lobe- but I only use that to make comparisons to spring forces of known-reliable engine combinations.

Jack,

Wise words indeed my friend.   There is a LOT of valve train stuff out there that is just poorly designed/poorly engineered.  As I have said many times before: "If your cam grinder can not or will not answer your questions, you need a different cam grinder."   Building reliable racing engines is a co-operative effort, and the cam(s) are only part of the equation.

For those who are following along and want a good basic book on camshafts/cam timing and other cam related aspects of performance/race engines, I recommend the following:

http://www.amazon.com/Choose-Camshafts-Maximum-Power-SpeedPro/dp/1903706599/ref=la_B0034P3Q6I_1_1?ie=UTF8&qid=1351632212&sr=1-1

It's a bit on the basic side, but none the less a good starting point.  His charts on cam duration to engine speed range ignore inlet & exhaust tract flow capabilities & engine flow demand, but are useful.
 
Fordboy

[Dean Los Angeles]

"Very simple and straightforward" . . . but way past my comprehension level.

Ah, it is very simple and straightforward. Please note the less complicated interpretation below.

All you have to do is determine the weight of the rocker, and where the center (beer) of mass is (neither of which is all that difficult), then hang the rocker on its side on a bifilar pendulum (more beer) of a known length, rotate the rocker slightly in the plane of its normal motion and (oh please, more beer) measure the oscillation rate.  From the results, a little calculation gives the moment about the CG, and another calculation (mmmmm . . . beer) corrects it to the center of the rocker shaft location.  Very simple and straightforward.

[Peter Jack]

 

Pete

[Stan Back]

But I'm a Red Whine guy.

[Freud]

Simplified to better describe the dude..........I'm a Whine Guy.

FREUD

[fordboy628]

Midget, et all,

In reply to requests for various formulas regarding racing/performance engine engineering, I am recommending the following books:

1   Performance Automotive Engine Math, by John Baechtel
2   Engine Airflow, by Harold Bettes
3   How to Choose Camshafts and Time Them for Maximum Power, by Des Hammill
 
These books are written in a more user friendly format for those who are NOT engineers.  There is however, a level of math intensive understanding (beyond basic math concepts) that is required for the information in these books to be translated into: useful information.

An additional caution is in order here.  Useful engineering requires valid input information.  That means no "guessing" about values.  You need to know the actual numbers for the formulas to work, not some bench racing guessing.  Have your heads flowed, your cam profiled, etc, etc, ad nauseum.   Then your numbers will be useful.

One last thought, about one third of my "bright ideas" work out, and my track record is pretty good.  Be prepared for the engineering to show that your combo/bright idea is NOT a winner.   My experience in racing is that most of the time, engine engineering is used to eliminate poor design choices, solve an error related to an educated guess, and/or most often, CORRECT A FATAL FLAW, AFTER THE FACT.........   Testing to destruction is the most expensive engine development program I know of, although there is an undeniable certainty about the results............

BTW, chilled, liquid, carbohydrate fueled "bench racing" is an activity I participate in all the time.  I've found some excellent micro-brews that way!!!
 
Fordboy

P.S.  If there are more of you who want a list of engineering level texts (or other texts) on this subject, send me a PM and I'll put together a list & post it up to Midget's Build Diary, or send it in a PM.