Keep cranks simple, or go high-tech?
From the February, 2009 issue of Circle Track
All contributors: David Vizard
Crankshaft technology has...
Crankshaft technology has evolved in both form and materials for the various professional racing series, even to the extent of creating significant power increases.
From the racer's point of view, a crankshaft's stroke would seem like the only dimension of any real importance. It may be, but we should not relegate the crank's other dimensional attributes to the realm of inconsequential. A crankshaft's No. 1 job is to convert the linear forces applied by combustion on the piston to rotational motion. All the power the engine is ever going to make is created above the piston crown. As a major part of the rotating assembly, a crank's job is to transmit power, as efficiently as possible, from the cylinder to the flywheel. The key words here are: as efficiently as possible. What we will examine in this article is how a crankshaft's configuration can affect its ability to efficiently convert linear power and motion to usable rotational power.
Efficiency First, let's look at crank efficiency. All too often, it is claimed that a long-stroke configuration will make more torque for a given displacement. The reason this is often quoted is that the engine has a longer lever arm with a long stroke. Although this may be true, it has sacrificed piston area, and that cancels out the stroke advantage. A longer stroke has more cylinder bore friction and limits the rod length. A long stroke signifies a shorter rod and greater friction-inducing side thrust on the cylinder wall. These factors cut output everywhere in the rpm range. These two factors alone tell us that, within limits, a short-stroke, big-bore, long-rod engine is the way to go.
Short-stroke cranks with a...
Short-stroke cranks with a longer rod are mechanically more efficient than short rods and a long stroke. For a given gas pressure (A), the force pushing the piston into the cylinder wall (B) is greater when rod angularity (C) goes up (as it does with shorter rods).
Windage For the budget-orientated classes, the rule book often states that the crank cannot have knife-edged or rounded leading edges for reduced windage. I'm sure many racers have wondered just what advantage an aero crank has over a square-faced, counterweight design with a stock pattern. Some years ago, the opportunity arose to run such a test in a 383 engine. Here, Scat 331/44-inch stroke cranks of each design (aero and regular) were used because the longer stroke would more clearly show what the difference was. Also, the oil used was 20-50 Mobil. That's probably a grade thicker than what might normally be used in a typical near-stock race engine. The results (Figure 1), though almost certainly showing bigger differences than would be seen in an engine with a shorter stroke, strongly indicate the advantage of an aero crank over a regular crank.
A point to note here is that if the crankcase has a vacuum drawn on it, the advantage of an aero crank diminishes. The results in Figure 1 tell us that aero cranks are good for power output.
Coatings About 25 years ago, coatings with various properties began to find their way into the top echelons of racing. Slowly, but surely, these have developed into one more weapon in the professional engine builder's speed arsenal. Thermal barriers for heads and intakes are a common feature of many race engines, but we are looking at cranks here. The ticket here is oil-shedding, Teflon(r) based coatings. Do they help power output? Working with Calico Coatings, I ran tests-again in a 383-and got the results, which are shown in Figure 2. Conclusion: Coatings deliver, but not until fairly high in the rpm range.
This Scat cast steel aero...
This Scat cast steel aero crank has been detailed with emery rolls and given an oil-shedding Teflon(r) based coating by Calico Coatings. Our dyno results showed an increase in top-end output.
Journal Diameter For about the last five years, with no-holds-barred race engines, the trend has been to reduce rod and main bearing journal diameters wherever and whenever possible. The obvious point of this is reduced frictional bearing loss. Just how much, you may ask, can that amount to? More than you might think, as bearing loss goes up much faster than diameter does. This tends to come down significantly faster as journal size is decreased. A move some years back was to drop the normal small-block, 2-inch, big-end journal size to 1.88 inches. Although the feedback numbers vary somewhat, it seems that a gain of 5 hp at 9,000 rpm is about average. But there is more to it than just the reduction of the bearing loss. The smaller big-end journal weighs less and, as a consequence, needs less counterweight to balance it. At first, you might think it's a direct trade-off, but that actually is not the case in most circumstances. Since the counterweight is spread out over a considerable arc, much of the weight is not fully effective. Consequently, it takes more weight to balance the effect of the more compactly located mass of the piston, rod, and big-end journal. The upside of this is that if the weight is removed from either end of the counterweight (instead of removing it right at the heavy spot), a lot more counterweight mass can be removed than is lost at the big-end journal side. By keeping a close watch on piston, rod, and journal mass, significant reductions in overall crank weight and moment of inertia (MOI) can be made. Since any short and medium circle track is substantially about accelerating off the turn, a lower MOI translates directly into better acceleration. Rear wheel dyno acceleration tests of high versus low MOI rotating assemblies show that efforts put into reducing the MOI of the crank, rod, and piston assembly can easily amount to another 10 rear-wheel horsepower. The faster the acceleration rate, the bigger the difference.
Bearings and Clearance Although coatings have had some acceptance as oil-shedding mediums, the place they have found most favor is in bearings. At this point in time, it would be true to say that probably better than 50 percent of the Cup car engines are utilizing coated bearings. Personally, I hardly ever build an engine these days without coated bearings, as they have shown to take the continued abuse of extended dyno sessions and come out almost unscathed. Exactly how coatings improve bearing life is not clear, but evidence indicates that coatings not only allow the oil to flow around the journal/bearing clearance volume more easily to maintain an intake film, but also have a small amount of surface porosity that causes the oil to soak into the coating just enough to effectively combat minimal lubrication conditions. Whether that is the case remains to be seen, but not knowing how they work certainly doesn't detract from the fact that they do work.
Reduced main bearing journal...
Reduced main bearing journal size means less weight, as do hollow journals. Contrary to what might be expected, a hollow journal with the correct geometry can deliver a greater fatigue life than a solid one.
With the need to minimize the amount of oil flying around inside the crankcase, efforts have been made to keep bearing clearances down to a minimum. The smaller the bearing clearances are, the thinner the oil needed to get the job done. This means taking far more care when measuring everything and making sure the clearances are accurate. The problem for the guy on a limited budget who is building engines at home is that measuring tools can be expensive. To establish bearing clearance, two pieces of measuring equipment are needed: an internal and an external micrometer. An external micrometer can be acquired at a very reasonable price from any mass tool and equipment merchandiser, such as Harbor Freight. What seems unavailable at any price under about $600 is an internal micrometer, which measures main bores and big-end rod journal bores. At a lower price, but still far from a giveaway, is a dial-bore gauge. Many engine shops use this, but it is not exactly easy to use. Give one to 10 engine builders inexperienced with their use, and you are likely to get a main bearing measurement at 10 different sizes. The internal micrometer, on the other hand, is about as close to a sure-fire deal as it comes, regardless of how much experience the builder has with this tool.
The difference between regular...
The difference between regular flat-faced counterweights (1) and aero counterweights (2) can be seen here. Dyno tests show aero makes more power.
Cranks break through the journal...
Cranks break through the journal radius (arrows). That is why grinding under size and making the radius bigger can actually strengthen a stock-sized crank. With a 1.88-inch journal diameter, the radii on this Cup car crank are far more critical.
So, where does this leave those of us building our engines at a home workshop with less-than-professional shop tools? Well, there is Plastigage. This stuff is about as cheap as it comes, but it is looked down upon by almost every Cup engine builder I've ever mentioned it to. Why? Because it is not as accurate as the more correct (and expensive) tools for the job. However, I think there may be an element of techno-social climbing going on here. Having used all three of the methods mentioned, here is my take on it. First, Plastigage accuracy is not as bad as some pros would have you believe. Testing this for yourself is easy. All you need is two machine parallels and two 0.002-inch feeler gauges. Just place the two feeler gauges side-by-side on one of the parallels and place a strip of the Plastigage between the feelers. Next, place the other parallel on top and squeeze the pair in a press or vice. When you measure, the now-spread Plastigage will have a reading that is really close to the 0.002 inch that it should read. The not-so-good news is that when applied to a curved bearing and journal, things are not quite that close. Generally, the results are within about +/-0.0002 inch (i.e., 2 ten-thousandths of an inch). Maybe this is not perfect, but it is good enough for most of us if the bearings are about middle limit. Now we come to the best part of using Plastigage. If the bearing being measured is out of tolerance, it will be obvious. If the bearing clearance is wrong, especially if it is on the tight side, it will instantly show up, thus preventing serious engine damage.
Now we know how to measure bearing clearance, but how much of it should we have? For most V-8s, a good working figure is 0.002 inch (2 thousandths) for the rod journals and 0.0025 inch (211/42 thousandths) for the mains. Going up half a thousandth on this is no big disaster. In fact, I have used as much as 0.004 inch (4 thousandths) on the mains when it has been a case of "use the crank or don't race." If you are building an engine that has good components, then these nominal figures can drop by half a thousandth (0.0005 inch). If you are building for a small-engined four-cylinder class, the clearances can also stand to be about half a thousandth (0.0005 inch) less than V-8 clearances.
The amount of crank-to-bearing clearance used also influences the weight of oil needed. The closer the clearances, the lighter (thinner) the oil needs to be. This is good news if you are reducing windage and crankcase scavenging losses are a priority.
Conclusions We have seen the advantage gained from coated aero cranks with reduced MOI in wet-sump engines, but there isn't necessarily a direct carryover to dry-sump engines pulling a lot of crankcase vacuum. The more vacuum pulled on the crankcase, the greater the tendency for oil to drop out of suspension. Sure, coated aero cranks are still an advantage for dry-sump engines, but not necessarily by such a margin. One direct carryover from wet sump to dry is the reduction in MOI. This is good, regardless of what type of engine you are running.
|Scat 331/44-inch Stroke Crank Test |
Stock Square-Face Counterweights vs. Aero Counterweights.
|RPM||TQ1||HP1||TQ2||HP2||TQ diff.||HP diff.|
|Figure 1: Shown here are the results of some very carefully run dyno tests. These numbers are the average of a substantial number of runs, with the best and worst of each discarded. By using this technique, the effects of the scatter of 2 or 3 lb-ft on each dyno run can be minimized. As the results indicate, there is a trend for the aero crank to show (as expected) a bigger increase with increasing rpm.|
|RPM||TQ1||HP1||TQ2||HP2||TQ diff.||HP diff.|
|Figure 2: Using the same test procedure as per Figure 1, these results were obtained from a fairly high-output, flat-tappet cammed, 383-inch small-block Chevy engine. The coatings appeared to be worth a reasonable amount of power in this wet-sump test engine, but not until engine speed had exceeded about 5,000 rpm.|
When choosing rods, don't...
When choosing rods, don't use any more rod than is necessary to manage the engine's output and rpm. These Lunati rods are an excellent compromise between weight and strength and are good for 750 hp in the 8,000-rpm range.
Reduced piston and rod weight,...
Reduced piston and rod weight, along with a hollow journal with a smaller diameter, allows the counterweight to be cut away much more to form a lighter pendulum counterbalance mass.
Because it concentrates the...
Because it concentrates the mass in a more favorable position, it is possible to use heavy metal slugs (arrow) to make a crank with less overall weight and a lower MOI.
Coated bearings are making...
Coated bearings are making a positive contribution toward longer life from smaller bearings.
A micrometer that has sufficient...
A micrometer that has sufficient accuracy to measure crank journals is easily within the budget of the average racer.
Acquiring a $600-plus internal...
Acquiring a $600-plus internal micrometer is usually well beyond the average racer's budget, but it is the way to go if you can afford it.
The strip of squashed Plastigage...
The strip of squashed Plastigage (arrow) is measured for width against the chart to establish the bearing clearance. Although not as accurate as expensive micrometers, Plastigage will reveal a potential failure situation due to incorrect clearances virtually every time.