There are several kinds of windage trays, but the basic concept is to reduce contact betwe
Blowin' In The Windage
Andrew Koppenhaver, via CarCraft.com: I'm a longtime reader and was happy to see the 347 stroker in the Mar. '10 issue. I'm not much of a Ford guy, but I am in the process of building a mild mill for my '78 F-100 and have been considering one of these stroker kits for some easy displacement. It was good to see an article that addressed some of my concerns and questions that have been stacking up in my head.
I'm writing this because of your final comment about using an aftermarket pan or windage tray. The article says in a comment by JMS, "The extra stroke tends to dip the counterweights into the oil, causing foaming and loss of horsepower." It's my understanding that when you grind more stroke into a crankshaft, it comes out of the rod journals, not the main journals, so that your crankshaft centerline stays the same. This being true, how can extra stroke cause the counterweights to dip into the oil? I could see the rod ends dipping in because of extra stroke, but not the counterweights. Can you shed some light on this? Also, isn't windage really the mist of oil present in the crankcase?
Jeff Smith: The comment could be termed an colloquialism, Andrew. Here's what's happening. Stroke is defined as the distance of the connecting rod journal centerline from the main journal centerline. The original Ford 5.0L engine used a 4.00-inch bore and a 3.00-inch stroke. To calculate displacement, use the formula bore x bore x stroke x 0.7854 x the number of cylinders (4 x 4 x 3 x 0.7854 x 8 = 301.6 ci). The 347ci engine bumps the stroke from 3.00 to 3.40 inches. This 0.400-inch stroke increase means the rod journals now extend another 0.200 inch farther away from the crankshaft main journal centerline, which puts the end of the rods that much closer to the bottom of the oil pan. This additional stroke really doesn't dip into the oil in the sump, but it does swing a wider arc, which means the ends of the rods will now be traveling at a greater velocity than a 3.00-inch stroke for the same engine rpm because the rods are traveling a greater distance.
You are correct that the term windage references the oil mist present in any engine. It's a combination of oil returning back to the oil pan and oil whipped by the spinning crankshaft counterweights. Oil level, sump depth, outside g-forces on the oil, windage trays, oil viscosity, oil temperature, bearing clearances, and probably a dozen other factors all play into this scene. The crank doesn't really dip into the oil level in the engine, but the counterweights are most certainly affected by having to travel through this oil mist. That's why most good performance crankshafts are tapered or knife-edged as opposed to having a blunt leading edge moving through the oil mist inside the crankcase. With a more aerodynamic shape, the crankshaft counterweight will expend less horsepower to travel through the oil mist at high engine speeds. At normal engine speeds-say less than 4,000 rpm-windage probably isn't a real issue. But with a longer stroke and engine speeds above 6,000 rpm, this becomes a significant point worth addressing.
We mounted a GTechPro Expandable Gauge System (EGS) tach on my orange Chevelle. This tach
Oxygen Sensor Stuff
Matt Kelley, Evansville, IN: Hey guys, thanks for a great magazine. I opted to get a cheap PC-based oscilloscope and an off-the-shelf narrow-band O2 to do some A/F tuning on my '65 Falcon since the oscilloscope would have other uses over a dedicated A/F monitor. I tried it out and it seems to work OK, but the response time for the O2 seems slow and the ratio seems to fall to super lean (0.2V) at cruise and even lower at idle. The Falcon has a 351W, Hooker full-length headers, an Edelbrock RPM cam, and a 650-cfm Holley on a Torker II intake all running through a T10 and some 3.50:1 gears. I have the O2 installed on the right bank as close to the collector as possible. Would the cam overlap produce the low voltage at idle? I'm probably not using the best O2, but would it cause the lazy readings? Any discussion on this or maybe a full article would be greatly appreciated.
Jeff Smith: Matt, you have an interesting idea to incorporate an oscilloscope with an oxygen sensor. The narrow-band sensors are just that-they will deliver an accurate voltage feedback for air/fuel ratio but only directly around stoichiometric or 14.7:1 air/fuel ratio (A/F). On either side of 14.7, the voltages are drastically different-roughly 0.85 for a rich mixture, while lean would be what you're seeing at roughly 0.2 volts or less. If you look at the illustration of the narrow-band output (see the accompanying graphs), the drastic voltage change becomes obvious.
A wide-band oxygen sensor delivers a much more linear voltage curve (the graph on the right side) that starts at very low voltage around 0.90 and extends through roughly 2.1 volts. These wide-band sensors use the reference voltage of 1.47 as stoichiometric (also called Lambda), which is 14.7:1 A/F. So voltage delivered from the sensor of 0.85 equates to 12.5:1 A/F (14.7 x 0.85 = 12.5). Given this simplified output, you can use a voltmeter to read the A/F on most wide-band oxygen sensors merely by moving the decimal point. So if the oxygen sensor output reads 1.25 volts, the A/F is 12.5:1.
This illustration shows the relative voltage output differences between a narrow-band sens
Concerning cam overlap, remember that oxygen sensors are aptly named since they don't measure the actual ratio of air to fuel. Instead, the sensor reads the amount of free oxygen in the exhaust and calculates an A/F ratio. This is why a misfiring engine will often read a leaner A/F ratio because the sensor is reading the free oxygen that is pumped through the engine when the cylinder did not fire. When we use a long-duration camshaft with additional overlap, both valves are open at the same time for a longer period of time. It is inevitable that exhaust gas and fresh inlet air will pass directly from the intake valve right out the exhaust. This free oxygen is read by the sensor as a lean A/F mixture when the truth is the A/F ratio is probably significantly richer. My experience with the Lester Scruggs 404ci LS engine in my orange Chevelle has the Innovate meter consistently delivering an A/F ratio at idle between 16:1 and 17:1 in gear with occasional spikes of 20:1. With the huge amount of overlap in the cam in this engine, I doubt the engine would even run at these A/F ratios.
As to your mention of response time-keep in mind that there may be two things going on here. One is the response time of the oscilloscope itself combined with that of the sensor. This is a complex subject that I'm not all that familiar with, but from what I've read, it seems that more expensive sensors tend to be much quicker than less expensive ones. The age of the sensor is also a critical factor. As the sensor ages, it will become slower and less responsive, which means the computer is always playing catch-up with the engine. This is one reason a new sensor can improve driveability.
Huntington Beach, CA
Tesla Electronics (GTechPro)
Santa Monica, CA
This is the brake pedal from my '67 Camaro. The hole closest to the top pivot hole (arrow
Pedal Travel Blues
Gary Tetu, Scottsdale, AZ: I love your magazine but wanted to let you know what happened after my professional shop installed a Wilwood 7/8-inch bore master cylinder in my '70 Plymouth Duster. My Duster has Wilwood disc brakes in front with drum brakes in the rear. I did not get the same impressive results you reported in the What's Your Problem section of your Mar. '10 issue. In the piece titled "Take a Brake," you reported a "very firm pedal" with "far less pedal effort necessary to generate the same braking force."
In my case, the brake pedal became soft and mushy and had 1.5 inches more pedal travel. The pedal effort required to generate the same braking force was only very, very slightly improved. My shop properly bled the brakes and tried multiple adjustments of the proportioning valve. It also tried a 10-pound residual pressure valve in the rear brake line to keep the rear brake shoes partially expanded and thus reduce brake pedal travel. To me the mushy brake pedal and increased pedal travel were just not worth it, so I had the shop reinstall my OEM 1-inch-bore master cylinder. Now I once again have a rock-hard brake pedal that requires little travel. So what if I have to push the pedal a tiny bit harder.
Just thought I would let you know my results. I wonder how other readers have or will make out with this modification. Based on my results, I hope they are not expecting this 7/8-inch bore master cylinder to feel like modern booster-assisted power brakes.
Jeff Smith: I should know by now that there are always exceptions, Gary. But that does not mean there isn't a way to allow you to have your cake and brake it, too. I've been spending a lot of time with my buddy Doug Norrdin at Global West Suspension, and in the middle of one of our many discussions on how to improve my '65 Chevelle's handling, I asked him to comment on your situation.
We actually ran into a similar situation on my '65 corner-turner Chevelle, moving from a 15/16-inch master to that same Wilwood 7/8-inch master cylinder you tried. When we had finished bleeding the brakes, I immediately noticed what felt like a soft brake pedal. We bled the brakes again but found no air in the system. What I was feeling was a change from the previous rock-hard brake to additional pedal travel (perhaps 1 inch-I didn't measure it) that was less than satisfying.
We talked about it and decided to test the car first before going back to the original master, which was showing signs of age. After posting the quickest overall time at the Run to the Coast Baer Brake Stop Challenge (see "Pro Touring Shootout," Aug. '10 CC), we decided that we could live with a little bit more pedal travel.
Doug's a pretty sharp guy, and he offered a couple of suggestions for your car. First, with all the air purged from the system, that mushy feeling is just the additional piston travel that is a direct result of reducing the area from a 1-inch to a 7/8-inch piston master cylinder. That's a huge change, but you don't have to live with that. Doug suggested looking under the dash to see if there may be two holes drilled in your brake pedal. On GM cars, the manufacturer used the same pedal for power brake- and manual brake-equipped cars, and there are two holes to attach the actuator rod. Doug says on average that power brake cars use a 5:1 pedal ratio while manual brake cars increase this ratio to 7:1. The closer the actuator hole is to the brake pedal pivot, the greater the pedal ratio. This ratio not only multiplies the force applied at the pedal but also affects the distance the pedal travels. The lower (power brake) hole not only decreases the force applied to the master cylinder but also increases the distance traveled.
I removed the power brake pedal from my '67 Camaro RS, measured the overall pedal length from the fulcrum to the center of the brake pad compared with the position of the two separate holes from the fulcrum, and then calculated the ratios. If the arm is 12 inches from the fulcrum to the center of the pedal and the uppermost hole is 2 inches from the fulcrum, this creates a 6:1 ratio. This is where the top hole is in my Camaro, but the bottom (power brake) hole measured 3.4 inches. That creates a much shorter 3.6:1 ratio.
Installing a manual 7/8-inch master cylinder on the car and using the lower (power brake) hole on the brake pedal would easily create a situation where the pedal travel is increased due both to the low ratio and to the smaller master cylinder piston diameter. Even with major leg force on the pedal, the pressure generated by the master cylinder would still be low because of the weak pedal ratio. To compute this difference, I used a 50-pound force applied to the pedal with a 6:1 ratio and calculated 499 psi of master cylinder line pressure. That same 50-pound force with a 3.6:1 ratio only generates 299 psi-a drastic 40 percent drop in pressure. In other words, you'd have to use 70 percent more pedal effort to create the same master cylinder hydraulic pressure.
It's possible that your Duster was fitted with power brakes from the factory and that the pedal ratio may be somewhere in the 5:1 range. If the pedal is not drilled for manual brakes, you can easily drill a 6:1 ratio hole. Change the pedal ratio and then bolt that Wilwood master cylinder back on the car. You should discover the pedal travel will be reduced and the hydraulic pressure will increase. You mentioned that the brake pressure measured at the calipers was only 800 psi, but I'll assume that was only with medium effort. With this smaller master and improved pedal ratio, the line pressure to the front brakes should increase to more like 1,100 to 1,200 psi. The pressure to the rear brakes will be less to prevent premature lockup.
Global West Suspension Components
San Bernardino, CA 909/890-0759
This is a typical diaphragm-style pressure plate with its Belleville spring (arrow). This
In The Clutch
Bill Snyder, Salt Lake City, UT: I have a '65 Impala with a 396 backed by a Muncie M21 four-speed. While I have the engine pulled, I am going to replace the clutch and pressure plate. This is a street car and will not be raced. I would assume there is a clutch package out there that is superior to an original setup-or not? What would you recommend?
Jeff Smith: There are many good clutch systems out there, Bill. Instead of giving you a specific clutch recommendation, let's talk about several I think would work for you, and you can choose the one that best fits. Roughly a half century ago, GM converted to what is now called a diaphragm-style clutch. The diaphragm uses a single Belleville-style spring that has an interesting advantage over three-finger (Borg & Beck or Long) clutches. These other designs use coil springs between the pressure plate and cover. Depressing the clutch pedal moves the levers, which compresses these springs. As with any coil spring, effort increases with pedal travel, which means your leg is pushing against maximum spring load with the pedal on the floor. That's why your left leg gets tired. An interesting by-product of the Belleville spring used in all diaphragm clutches is that as the fingers of the spring are compressed past a certain point, spring pressure is drastically reduced. This means that when the clutch pedal is fully depressed (at a stoplight for example) your leg is only working against a fraction of the spring's total clamp load. This is why diaphragm clutches are almost universally used in street applications.
Now that we've established the advantage to diaphragm clutches, all we're left with is choosing the model for your Impala. Your big-block requires an 11-inch clutch and pressure plate with a 10-spline input shaft with the Muncie. I looked up kits from McLeod (StreetPro PN 75124, Summit Racing, $256.69), Hays (PN 85-110, Summit Racing, $259.95), and Centerforce (Dual Friction, PN 73552, Summit Racing $269.95). All three of these kits include an 11-inch disc and diaphragm pressure plate. Most also include a plastic clutch alignment tool, and the Hays and McLeod kits also include a new throwout bearing, which should always be changed when installing a new clutch and pressure plate. That makes the Centerforce a little more money, but it incorporates a Dual Friction clutch disc, which offers a little more holding power compared with the other two organic discs. You really can't go wrong with any of these clutch kits.
We didn't talk about flywheels, since I'm assuming you will retain your stock one. But always have the flywheel surfaced whenever adding a new clutch and pressure plate. Another important step is a new pilot bushing to ensure that the transmission input shaft is aligned with the crankshaft. If the clutch kit you purchase does not include the hardware, then be sure to purchase new quality pressure plate bolts. We also use a drop of thread-locking compound on the flywheel bolts when torquing them in place. Inspect the clutch fork and especially the ball stud in the bellhousing for wear. The ball is often overlooked and can cause a side load on the clutch fork that can cause wear on the throwout bearing and the transmission input shaft collar. Look everything over very carefully, including all the clutch linkage pivot points. These are often badly worn. Remember, we're dealing with cars that are now almost 50 years old. Choose your clutch and enjoy how smoothly it will work.
Mr. Gasket (Hays)
Also at the Knott's Berry Farm Show was John Vermeersch's '61 Ford wood-side wagon pulling his boat that just happens to be powered by a Roots-blown SOHC 427 Ford motor. John is the longtime owner of Total Performance in Mount Clemens, Michigan, and a lifelong Ford fanatic. Car Craft did a cruisin' story with him and his SOHC-powered Starliner back in the '80s.
Looking straight down into a dual-plane intake manifold like this Weiand Stealth Air Strik
Playing With Plenums
Otis Linton, Santa Maria, CA: Stories have been told about cutting out the divider that splits the plenum in the intake manifold on a small-block Chevy. Have you ever tested this? I've always wanted to try it.
Jeff Smith: Don't you just love a guy who comes right to the point? Let's see if I can be as parsimonious with an answer. The divider is the vertical wall that separates the two small plenums on a dual-plane intake manifold. This creates two separate intake tracks that split the engine into a pair of four-cylinder engines. On a small-block Chevy the cylinders are separated 90 degrees apart by the firing order. With a firing order of 1-8-4-3-6-5-7-2, this combines cylinders 1 - 4 - 6 - 7 together in one plenum and 8 - 3 - 5 - 2 on the other. Looking down through the top of the manifold with the carburetor removed, the first thing that becomes apparent is that each plenum is rather small and that each half is fed by one primary and one secondary barrel on a typical four-barrel carburetor.
The advantage of a dual-plane manifold is that the intake runners are significantly longer than a single-plane. Longer runners boost the midrange power-roughly between 3,500 and 5,000 rpm. A single-plane intake employs much shorter runners, which emphasizes top-end power at the expense of midrange torque. These are general characteristics that can and do overlap depending on individual intake design. One disadvantage of the dual-plane is that one plenum floor tends to be higher than the other, creating a smaller common (plenum) area. At greater engine rpm, the high-speed air and fuel column through the carburetor tends to crash into this shallow floor, disrupting flow and causing air/fuel separation. This then turns some fuel back into a liquid, which is much harder to burn. This same situation can occur with a single-plane intake manifold, which is why open-plenum carb spacers tend to make more peak horsepower on some manifolds. Adding height between the bottom of the carburetor base plate and the floor of the manifold gives the air and fuel a much gentler radius to make the 90-degree turn into the intake ports. But we digress.
Removing the divider wall between the two small plenums in a dual-plane intake manifold first allows the two plenums to communicate with a larger common area from which to pull air and fuel, which improves high-speed power. Often removing this wall will help power everywhere in the entire engine rpm operating range. Other times, cutting this wall will help the top-end power but hurt the low-speed torque. One way you could test this before modifying the intake manifold is to add a 1/2- or 1-inch-tall open-plenum spacer to the intake manifold (assuming there is sufficient vertical room to the hood when the spacer is added). Adding an open plenum spacer will also affect the fuel delivery curve, so that's one thing to watch carefully. Adding a spacer tends to lean the fuel curve because the signal to the carburetor has been reduced slightly. If the carb is already on the rich side, an increase in power may be due to leaning out the air/fuel ratio slightly-or some combination of manifold design change and air/fuel ratio change. You can see why testing all this stuff can get complicated.
Cammers Are Cool
We can never get enough of the Ford Cammers. This one was at the 25th anniversary Knott's Berry Farm Show. Stuffed into a '64 Fairlane T-Bolt tribute car, the motor displaces 440 ci with a Velasco steel crank, 14:1-compression Arias pistons, and that iconic Hilborn injector stack setup. The car is a tribute to Emmett "Rattlesnake" Austin.
Here are Comp Cams' new Ultra Pro Magnum rocker arms. They are investment-cast 8650 chrome
Greg Smith, Woodland Hills, CA: I have a stout small-block in a '55 Chevy with an aggressive mechanical roller cam that has been in the car now for years. I don't put a lot of miles on the car, but when I do, it's making quarter-mile passes on a regular basis. Last weekend I wanted to check the lash. The cam is a custom grind from Comp with 264/269 degrees of duration at 0.050-inch tappet lift with 0.631-inch valve lift on the intake and exhaust with a 112-degree lobe-separation angle. My Comp cam card specifies 0.026 and 0.028 inch of lash, but it does not say whether it should be set cold or hot. If I set the lash with the engine cold, does the clearance increase or decrease after the engine is warmed up? I've received two different opinions from friends.
Jeff Smith: Great question, Greg. We decided to go to the source, so I called Billy Godbold, Comp Cams' rocket scientist turned cam designer. First, Comp always specs lash as a hot setting. This is probably true of all cam companies because lash is designed to accommodate changes in engine clearances due to thermal expansion. Billy says lash loosens on all pushrod engines as the engine warms up. In general, he says an engine with an iron block and heads will increase the lash about 0.003 to 0.005 inch. An iron-block engine with aluminum heads will grow roughly 0.005 to 0.007 inch, and all-aluminum engines grow the most-anywhere from 0.007 to 0.015 inch or more.
Let's say your new engine has an iron block and aluminum heads and has never been started. Billy recommends setting the lash at 0.020 inch cold, which allows more than enough clearance for the exhaust valve to fully close as the engine warms up. Then recheck the lash after the engine has achieved a normalized temperature. If the lash increases about 0.005 inch for a total of 0.025 inch, this is very close to Comp's hot recommendations and you can adjust accordingly. Billy says you should never try to run too tight a lash when cold-nothing less than 0.015 inch. This is because if the engine ever saw very cold temperatures, it's possible the exhaust valves would not close and would burn up during startup before the engine achieved operating temperature.
This also brings up the concept of how much the recommended lash can be altered. Billy says most flat-tappet and roller cams with lash specs around 0.020 inch can be run as tight as 0.016 inch and as loose as 0.030 to 0.035 inch. The only reason to change the lash would be to determine what effect a change in duration would have on the engine. As an example, Billy likes to work on the intake first since it tends to have more of an effect on engine performance. He prefers to work in stages of 0.004 inch by loosening the lash and then testing the engine. Loosening the lash effectively shortens the duration by roughly 2 degrees. If the power increases (more trap speed in dragstrip testing), it indicates the camshaft duration is too long by at least 2 degrees. In this case, you would add another 0.004 inch of lash and run the car again. In the case of your camshaft, this has now pushed the lash up to 0.034 inch while cutting the duration by roughly 4 degrees. Keep in mind that loosening the lash also reduces lift, but 0.008 inch isn't that much. If the dragstrip trap speed improves again, we would suggest that shortening the duration tends to improve midrange torque perhaps at the cost of top-end power. A trap speed improvement in this case is probably due to the increased torque accelerating the car more efficiently as evidenced by a higher trap speed and perhaps a lower e.t.
While we're on the subject of lash, this offers the opportunity to smash a popularly held performance myth about street-driven mechanical camshafts. It's common to hear someone say, "I don't want a mechanical cam because I don't like setting lash all the time." I think this started with '60s small-block Chevys where the factory used a 3/8-inch fine thread pinch nut. These locking nuts work great-once. After only a couple of adjustments, the nut loses lock tension, it loosens, and the frequent adjustments begin. Today, the smart move is to use a simple poly lock. They don't move if you adjust them properly. As a reality check, I have a John Lingenfelter-built 420ci small-block in my '65 Chevelle that has been in the car since about 1990. Over the course of 20 years of abuse, the lash has changed perhaps 0.002 inch, and I'll attribute that more to how I estimate the pull on the feeler gauge. I still check the lash about once a year just to make sure something isn't wearing out, but it never changes. If the lash changes, something is wearing, bending, or about to break.
The bottom line is lash will not change if everything is happy.
Given that nasty Kaase Boss in this issue, we thought we'd continue the Ford flavor with Rick Stanton's '69 Talladega powered by a 598ci Boss motor. He stroked his Boss with a 4.50-inch crank, and at Westech it tweaked the dyno to 824 hp at 7,600 rpm. With an Isky roller cam and a 1,250-cfm Dominator carb, this beast smokes.