In Part 2 of Finding Balance, we covered some of the variables that go accounted for during most traditional engine balancing processes. The million dollar question is:
Given all these variables, how can engine builders
improve upon standard balancing techniques in high-rpm, big horsepower applications?
Racers and industry experts have been attempting to solve this riddle for decades. Through a combination of incredibly advanced technology and good old fashioned trial and error, engine builder are making steady progress in navigating these murky waters.
Since GM, Ford, and Chrysler (or any other OE manufacturer) have resources far beyond that of even a top-tier engine shop, it’s not surprising that they have developed incredibly advanced tools to assist with the balancing process. To eliminate all the variables that a standard balancer can’t measure, the world’s ultimate balancer would allow running an engine on the dyno at idle, part-throttle, and WOT. The OEs aren’t there quite yet, but GM has taken a stop in the right direction.
At the GM LS7 engine assembly plant, the company utilizes dynos that spin the motor to approximately 2,000 to 3,000 rpm. A series of sensors attached to the motor measure the imbalance, and a technician then adds or removes mass from the flywheel as necessary. While GM performs an initial baseline balance before the engine is assembled, this final step allows fine tuning the balance beyond what’s possible using standard balancing equipment.
It’s important to note that the engine isn’t running under its own power during this balancing process. Instead, it’s the dyno that’s rotating the engine over. While this makes it impossible to directly measure the effects of cylinder pressure during the power and exhaust strokes, the dyno can still replicate the forces generated during the intake and compression strokes. Equally as important is the dyno’s ability to measure the effects of ring drag, bearing friction, counterweight phasing, rod-to-stroke ratio, and static weight on balancing.
Considering that the counterweights are the components that are actually responsible for balancing the rotating and reciprocating forces inside an engine, logic says that refining the balancing process starts with the counterweights. When comparing an off-the-shelf cast or forged crankshaft to a custom billet crank, engine builder often attribute reduced engine vibrations and bearing wear to reduced crank flex and the addition of the center counterweights. These are valid points that shouldn’t be ignored, but it’s also possible that much of the improvements are attributable to the actual positioning of the counterweights in relation to the rod journals.
Counterweight positioning, or phasing, is something that’s seldom discussed, so it’s a term that probably doesn’t register with most people. To get a better handle on this concept, imagine looking at a crank from the head-on profile with the snout up front and the rear main in the back. The counterweight phasing is simply where the counterweight is positioned in relation to each rod journal. While most literature on balancing states that the counterweights are positioned 180-degree out from the rod journal, no one has bothered defining what 180-degrees out actually means.
For example, if the entire mass the counterweight was moved clockwise by several degrees, most of its mass would still be 180-degrees out. Likewise, if the entire mass the counterweight was moved counter-clockwise by several degrees, most of its mass would still be 180-degrees out. Obviously, when carving out a crankshaft from a single piece of billet steel, custom crank manufacturers like Bryant or Windberg can position the counterweights wherever they want in order to experiment with its effects on balancing. Companies on the cutting edge of technology usually aren’t willing to disclose their trade secrets, so whether they determine counterweight phasing based on mathematical formulas, trial and error, or both is a mystery.
Of course, all this talk about counterweight phasing assumes that it has positive effects on crank balancing in the first place. That said, the negative effects of positioning the counterweights in the wrong place have clearly been demonstrated in real-world testing. In extreme situations, the positioning of the counterweights can be so far off that even after balancing to ¼ ounce-inch, a motor can rattle itself to death and destroy the bearings. Although this doesn’t happen often, every now and then a defective crank will slip past a manufacturer’s Quality Control department, and the carnage they create leaves people scratching their heads. I’ve personally witnessed this happen on the dyno.
If poorly placed counterweights can wreak havoc on engine balancing, this seems to suggest that ideally placed counterweights will even out bearing loads and prolong engine durability. Evidence certainly suggests that this may be the case. Although it flew under the radar at the time, Ford experimented with phasing the counterweights in different positions during the ‘60s in its Le Mans road racing engines.
Similarly, one of my graduates that works for a NASCAR Sprint Cup team was tasked with carrying out a similar experiment. In it, he removed material from the leading edge of the counterweights, and added the same amount of material back onto the trailing edge of the counterweights. This effectively changes the counterweight phasing in relation to the rod journal. Not surprisingly, my anonymous source wasn’t at liberty to discuss the results of this experiment, but the amount of time and money one of the top engine shops in the country was willing to invest in this research suggests that the potential benefits of phasing the counterweights just right are significant.
Based on the aforementioned examples drawing any definitive conclusions regarding the effects of counterweight phasing on engine balance isn’t possible. It’s hearsay at best. Fortunately, engineers at North Carolina State University conducted extensive lab experiments to test this theory in the late ‘90s on a NASCAR Sprint Cup small block Chevy. The results can be read in SAE paper 960354, and the results are eye-opening to say the least. By changing the position of the counterweights while leaving their overall mass relatively unchanged, the engineering team of Robert Sharpe, J.W. David, and Erik Lowndes reduced average bearing loads by 74 percent, and peak bearing loads by 41 percent.
The primary focus of the lab testing was to test the effects of counterweight phasing on engine balance while also accounting for cylinder pressure. During the experiment, the overall mass of the counterweights was left relatively unchanged, but the center of mass—and therefore the location of the counterweights—were optimized to reduce bearing loads. To simulate the operating range in a typical Cup race, the testing procedure varied engine rpm between 5,200 and 7,200. The key piece of data gathered during the test was comparing the bearing loads generated with an unmodified racing crankshaft to the loads generated with the modified crankshaft. Since bearing loads peaked at 6,200 rpm, the following figures were measured at that point.
With the unmodified crank, the baseline average bearing loads on main caps #1 through #5 read 19.89-, 24.20-, 10.94-, 27.72-, and 22.65 kilonewtons of force, respectively. Keep in mind that one kN equals 240.8 pounds of force, which is significant to say the least. With the modified crankshaft, those readings plummeted to 5.97-, 9.76-, 10.90-, 9.96-, and 6.08 kN, respectively. While the load on bearing # 3 didn’t change much, the load on all the other bearings was reduced 60 to 74 percent.
The reduction in maximum bearing load was extremely impressive as well. With the unmodified crankshaft, the baseline peak bearing loads on main caps #1 through #5 registered 38.59-, 45.15-, 27.13-, 48.75-, and 41.60 kN, respectively. As expected, main caps #2 and #4 experienced substantially greater loads, which isn’t surprising since bearings #2 and #4 typically experience the most wear. After modifying the crankshaft, the bearing loads dropped to 25.40-, 28.15-, 26.40-, 28.62-, and 25.62 kN, respectively. With the exception of the center bearing, this represents a 34 to 41 percent reduction in peak bearing loads. Equally as revealing is how effectively re-positioning the counterweights evened out the bearing loads across all five main caps. With the modified crank, bearings #2 and #4 are no longer taking the brunt of the abuse, which can only lead to increased crankshaft longevity.
Although the cited SAE paper doesn’t go into great detail on exactly where the counterweights were re-positioned to, which could have been intentional, the changes in bearing loads are staggering. Equally as surprising is the fact that the center of mass of the counterweights was only moved a few millimeters to optimize crank balancing. This begs the question, why does altering counterweight phasing affect overall balance and bearing loads so profoundly? Crank flex could be the key.
There’s lots of talk about how balancing affects crank flex, but very little talk about how crank flex might affect balancing. Consider that, when measured from front to back, a crankshaft can flex quite few degrees. It’s already been established that changing the counterweight positioning just a few millimeters can dramatically affect balance. That said, if crank flex caused the position of the rod journal to change in relation to the position of the counterweight, it’s conceivable that this flex would have some degree of impact—or perhaps even a very large impact—on balance.
Even if a crank did not flex between the rod journal and counterweight, or if this flex was too insignificant to affect balance, the crank can still flex torsionally along its own centerline. In this scenario, the location of the rod journals in relation to one another would change. For example, instead of rod journals number 1 and 2 being positioned 90 degrees apart, crank flex could squeeze this dimension together just a tiny bit to 87 degrees, or spread it apart to 93 degrees.
Minor flexing of just a few degrees in either direction may seem inconsequential. However, the fact that secondary balance is dependent upon maintaining consistent 90-degree phasing between the rod journals cannot be overlooked. Flex along the crankshaft’s rotational axis could potentially compromise a crossplane V8s ability to naturally cancel out secondary vibrations. Yet another factor to consider is that if this type of flex did in fact upset the secondary balance, a shorter connecting rod could conceivably make things worse since it would accelerate the piston away from TDC even more quickly than a longer rod.
Perhaps someday technology will exist that allows bolting a running engine to a balancer under all operating condition, thus eliminating nearly all the variables that can affect engine balance. Until then, modifying counterweight phasing seems to be one of the most effective methods of optimizing balance in a high-rpm racing engine.
Has anyone tried progressively balancing a crank light to heavy from front to back i ask this as years ago i ground and smoothed a crank in a racing straight six Holden engine that is not fully counterweighted. they are renowned for torsional twist and all vibrate around 6200 6300 i ground weight from the crank progressively from front to back had it balanced with a medium weight flywheel, a Romac balancer and that engine did not vibrate anywhere from 1000 to 8500 curious ?
I should add the weight removal was from all over especially the leading edge of the counter weights that it does have.
[…] Finding Balance (Part 3): Getting Beyond the Shortcomings of Traditional Engine Balancing […]
[…] In Part Three, we’ll take a look at potential solutions for the shortcomings we listed. […]