method for balancing/weighting a crank shaft?

126 views
Skip to first unread message

Dorkmo

unread,
Dec 9, 2011, 10:20:01 PM12/9/11
to Open Source Steam
so whats the best way to calculate add and weight to a shaft? do you
add exactly the amount that the piston and connecting rod weigh or is
there something else?

Ken Helmick

unread,
Dec 9, 2011, 11:43:18 PM12/9/11
to open-sou...@googlegroups.com
Wow,
 
Actually, that's my primary job at GM, balancing crankshafts, flywheels, dyno components and so on for all the stuff that moves through Engineering World Headquarters.
 
You could write a book on this subject, people have, actually....I've written a few really long articles.  You can find one of them at:
 
 
and scroll to the bottom of the page and download the file Engine_Balance_2003.pdf.
 
The general rule of thumb, if balance shafts are not involved, is that you need to attach a weight to each crankpin equal to the rotating masses attached to that pin, plus half the reciprocating masses.
 
NOTE:  You can NOT dynamically balance a crankshaft without specialized equipment, dynamic balance being a property of objects that are long enough in relation to their diameter to possess significant rocking couples.  You can balance a single cylinder crank acceptably enough, if you are careful and the crank is carefully built and symmetrical to begin with.  If you plan on rebalancing the LS-9 in your ZR-1 Corvette, please put my name down as beneficiary.
Anyhoo....the reciprocating masses would be the pistons, rings, wrist pin, wrist pin bushing and so on.  The rotating mass would be the crankpin bearing or bushing.  The connecting rod is where things get hinky, the big end makes a circle because it is on the crank and the small end goes uppy-downy in the cylinder bore.  It's one of those situations where the answer is "YES" if asked whether the rod is reciprocating or rotating mass. 
 
What we end up doing is weighing each end of the crank to properly allocate rotating and reciprocating masses.  Of course, being in the business and all that we have a sooper dooper special custom built scale that does the job instantly.  That's nice but not necessary, just cuts down on wasted time.  What you do is get an electronic scale and put a couple of spacers on top along with a steel pin, and then zero out the scale.  You set up another pair of spacers next to the scale, and also have a steel pin ready.  The rod is then suspended from the spacers, through the big and little ends of the rod, by the pins.  The spacers should be such that the rod is parallel to the ground.  If the big end is on the scale the weight on the readout is equal to the rotating portion of the rods mass, if the little end is on the scale, it is reading the reciprocating portion.  Now flip the rod end for end and repeat.  Now, as a double check, weigh the rod and see if the total weight equals the reciprocating and rotating weights.  Repeat as necessary until you get a solution that jives.
 
Now that you have the reciprocating and rotating masses summed up, you need to make some sort of weight to put on the crank pin.  This weight will be equal to 1/2 the recipocating mass plus the rotating mass.  There is a logical explanation for that and you can find it in the pdf file mentioned above.  We refer to this as a 'bob weight' or a 'ring weight'.  Most speed shops use universal weights to which they can add extra mass equally to each side until getting the right weight.  I make them on a lathe to fit the crankpin diameter, and also to have no more than .001 inch side play so the weight can not move fore and aft to change the load distribution AND I also statically balance the weight on a flywheel balancer when I am adjusting the weight to the desired value so as to guarantee that the center of gravity is dead in the middle of the weight.  For all practical purposes the crank sees the weight as a point source located at the CG, if the CG is not coaxial to the crank pin then the position of the weight (and the mechanical advantage) shifts depending on what angle the weight is bolted onto the crank.
 
Now, the old timers building small steam engines calculated the weights and measured out a long hank of plumbers lead until they got a piece that was the correct weight.  Then they would slowly and meticulously wrap it on the crank pin taking painstaking care to lay the winds on tightly and evenly so as to achieve the most equal distribution around the pin as possible.  This is hardly up to racing or factory specs, but I have done that for a few single cylinder test engines when time didn't permit constructing a special weight and the universal weight set I keep for emergencies didn't go small enough.  Since they weren't pushing the test mills to 8,000 rpm, no one saw any problems with the balance.
 
Have I scared you off yet?
 
:-D
 
Ken

Dorkmo

unread,
Mar 3, 2012, 3:41:49 AM3/3/12
to open-sou...@googlegroups.com
hey ken,

thanks so much for the detailed response. ive been slowly trying to get a grasp of it all and what changes i need to do.

ive been putting this off for while but im going to try and figure it all out this week


used the volumes of each piece from the sketchup model and got weight with a basic steel density number.

my main goal is to be able to make this simply, theoretically with a hack saw and a welder, forfeiting some rpms to lack of precision i suppose.

ill post some drawings of what i come up with. maybe well get a good laugh out of it.

and yes, im scared :O

Ken

unread,
Mar 3, 2012, 10:45:43 PM3/3/12
to Open Source Steam


I look forward to seeing what you come up with, I'd never laugh at
someone who is taking the effort to research the topic and taking
pains to develop a rational plan of action based on that research.
When someone does little research, assumes almost instant command of
the topic and then proceeds to repeat the classic errors that have
been debunked for generations....then I tend to laugh.

Your spreadsheet looks about right. When calculating the masses to
design the crank and counterweight, it is vital to know the location
of the centers of gravity. Two equally heavy counterweights respond
very differently depending on how far the center of gravity is from
the axis of rotation.

Counterweights are often strangely shaped, and it's a bit difficult to
determine what their weight and CG will be beforehand. If you have
access to better CAD programs, this information is at your
fingertips. The old timers sometimes got very close with drafting
tools, a knife, scale and construction paper. They would draw out the
counterweight on a piece of construction paper and cut it out very
carefully, then weigh the paper on a precision scale. Then they would
cut out a square from the same material and weigh it. If the paper
square was 5 inches on a side, the area was 25 square inches. If the
weight was 2.5 ounces, then the weight was 0.10 ounces per square
inch. From this point it was easy to determine the area of the
counterweight and by multiplying by thickness obtain volume. The same
heavy paper model was good for finding center of gravity. You would
block a metal rule edge up on the table, lay the paper cutout on it
and move it back and forth til you got it to balance. If the part was
symmetrical, by carefully marking the line described by the rule, you
could find the CG. If not too symmetrical, you would rotate the part
90 degrees on the rule and balance again, the intersection being the
CG. To determine the unbalance force, you then merely had to measure
the distance between the CG and the center of rotation and multiply by
the part weight.

OK, I admit these are kind of crude methods in the computer age, but
perfectly suited for a hacksaw and welder.

Regards,

Ken


On Mar 3, 3:41 am, Dorkmo <dor...@gmail.com> wrote:
> hey ken,
>
> thanks so much for the detailed response. ive been slowly trying to get a
> grasp of it all and what changes i need to do.
>
> ive been putting this off for while but im going to try and figure it all
> out this week
>
> i started a spreadsheet here:https://docs.google.com/spreadsheet/ccc?key=0AgxMMqGvwTM-dEswZUZ6d1hR...

Dorkmo

unread,
Mar 8, 2012, 10:55:02 PM3/8/12
to open-sou...@googlegroups.com
k im finding that im both unable to put enough weight within the original circle and my center of gravity of the weight is closer to the axis than the pin.

so i think my only option is to make the circle larger and extend on the weighted side?

i still need to go check and see what the maximum room i can make inside related to everythign else.

i put the image on canv.as if anyone wants to sketch on the pic real quick  http://canv.as/p/otyhw just a test to see how that goes
piccrank.png

Dorkmo

unread,
Mar 8, 2012, 10:58:20 PM3/8/12
to open-sou...@googlegroups.com
k theres some room to play with. max radius is 4-7/16" without any comfort zone.

Dorkmo

unread,
Mar 8, 2012, 11:00:21 PM3/8/12
to open-sou...@googlegroups.com
oops make that 4" i forgot to measure to the sides :O

Ken

unread,
Mar 9, 2012, 6:38:41 AM3/9/12
to Open Source Steam
It isn't the counterweight center of gravity, per se, that is
critical; it is the moment arm that is important. The moment arm is
the weight multiplied by the distance of the CG to the center of
rotation.

In the English system, we typically use "ounce-inches" to measure
imbalance, not suprisingly this being a measure of torque. Assume we
have a crankpin at 1.5 inches from the main bearings, a rotating mass
of 11 ounces and a reciprocating mass of 18 ounces. Our ringweight
would be equal to the rotating mass plus half the reciprocating, or 11
+ (18/2) = 20 ounces. Since the center of gravity of the ringweight
would be right on the center of the crankpin, the moment of the
ringweight would be 20 ounces x 1.5 inches = 30 ounce-inches.

Our crank will need to be heavy by 30 ounce-inches in the direction
opposite the crankpin, so that it is balanced when the ringweight is
added. The precise center of gravity isn't important so long as the
weight multiplied by the distance to the CG equals 30 ounce-inches.
If the distance to the CG is 1.25, then a weight of 24 ounces will be
needed (24 x 1.25 =30).

Note that while I have talked about the ring weights, the designer
also has to take the mass distribution of the crank itself into
account....the crankpin has a moment arm that must be offset, for
example.

Regards,

Ken

Dorkmo

unread,
Mar 17, 2012, 2:00:38 AM3/17/12
to open-sou...@googlegroups.com
okay.. i moved some stuff around and dumbed down the shapes a bit. gota finish the other parts so i can weight them.

from the estimates on the spreadsheet the moment of the ringweight is about 73oz-in
38.85 oz x 1.875 in

i made the welded on weight 2x3 with a 1x2 triangle cut off the end. then i moved the pin further down to make an inch at the other end work as weight. turned out as 12 square inches.

ended up with 80.5oz-in
40.896 oz x 1.96875 in

im a bit off but i figure i shouldnt stress too much till i get actual measurements.

i think its going to be another 3-4 weeks before i get closer :/

also found this handy centroid calculator for sketchup:
crankweights.jpg

Dorkmo

unread,
Mar 17, 2012, 2:06:05 AM3/17/12
to open-sou...@googlegroups.com
it was funny last week i was out in the oil patch and pumping unit weights kinda clicked. theyre kinda interesting bc they seesaw so the crank arm is attached to the other side where the weights are. or something like that lol maybe i just think i understand.

Ken Helmick

unread,
Mar 21, 2012, 8:59:38 AM3/21/12
to open-sou...@googlegroups.com
Sounds like you are on the right track, anyhow!  A centroid calculator is exactly the kind of tool the serious designer uses, though if they are richer than me they can design the entire crank in 3D and have the software calculate all the moments....cheaters!
 
Ken



 
-----Original Message-----
From: Dorkmo <dor...@gmail.com>
Reply all
Reply to author
Forward
0 new messages