The Basics of Wet Flow Cylinder Head Testing – Part 4

When you throw a lit match onto a gasoline-soaked rag, it ignites very easily. This is not an experiment I recommend or that is even needed. However, when gas so easily combusts, it’s hard to imagine there ever being a problem igniting it in the cylinders of a hot, high-compression engine. The reality, though, is that we have all the time in the world to wait for a burning rag to exhaust its fuel through combustion. In a well-developed engine turning 9,000 rpm, the time for 90 percent of the charge to burn is one and-a-half thousandths of second!


sa215smallThis Tech Tip is From the Full Book, DAVID VIZARD’S HOW TO PORT & FLOW TEST CYLINDER HEADS. For a comprehensive guide on this entire subject you can visit this link:
LEARN MORE ABOUT THIS BOOK HERE

 

SHARE THIS ARTICLE: Please feel free to share this post on Facebook / Twitter / Google+ or any automotive Forums or blogs you read. You can use the social sharing buttons to the left, or copy and paste the website link: http://www.musclecardiy.com/cylinder-heads/the-basics-of-wet-flow-cylinder-head-testing-part-4/



Fig. 4.1. Here is one of my hot street small-block Chevy engines. This one displaces 383 ci (6.28 liters) and dyno’d out at 609 hp (97 hp per liter) and 538 ft-lbs (85.7 ft-lbs per liter). This sort of output for a pump-gas 10.5:1 street-drivable motor of reasonable cost means covering every possible aspect of the power production chain of events. Good mixture quality and distribution are as vital as good airflow.

Fig. 4.1. Here is one of my hot street small-block Chevy engines. This one displaces 383 ci (6.28 liters) and dyno’d out at 609 hp (97 hp per liter) and 538 ft-lbs (85.7 ft-lbs per liter). This sort of output for a pump-gas 10.5:1 street-drivable motor of reasonable cost means covering every possible aspect of the power production chain of events. Good mixture quality and distribution are as vital as good airflow.

Wet-Flow Testing— What’s it Worth?

During my Mini Cooper era in the early 1960s, I first realized that the fuel droplet size and the state of the mixture quality (or lack of it) dramatically affects power output. I learned a lot on the A-Series engine that powered this car—much of it totally contradictory to what you might expect. If the engine that powered these race-dominating roller skates was fed with fuel even a little too finely atomized, the power dropped like a rock. The valuable lesson I learned was to never take anything for granted.


Fig. 4.2. Seen here is the late Joe Mondello showing off his wet-flow attachment for a conventional Super Flow bench. This add-on is available for most of the commercially available race shop benches on the market.

Fig. 4.2. Seen here is the late Joe Mondello showing off his wet-flow attachment for a conventional Super Flow bench. This add-on is available for most of the commercially available race shop benches on the market.

 

Fig. 4.3. Dart’s wet-flow bench is truly a monster. Unlike most of the wet-flow test systems, it combines flow capability in CFM and the ability to analyze wet flow patterns. Note the huge pumps and electric motors needed for this bench to perform its various functions.

Fig. 4.3. Dart’s wet-flow bench is truly a monster. Unlike most of the wet-flow test systems, it combines flow capability in CFM and the ability to analyze wet flow patterns. Note the huge pumps and electric motors needed for this bench to perform its various functions.

This paid off handsomely a few years later (1972–1975) when, after winning a drag race championship with the same type of engine, I ended up developing a dominant four-cylinder Chrysler engine for the British Touring Car Championship series in England. With this engine the route to success was exactly the reverse of my success with the Mini Cooper’s engine. Very finely atomized fuel and good dispersion, along with very low re-aggregation, were the key factors. They allowed me to produce a carb/intake manifold and head combination that substantially out-horse powered (by about 20 hp) the two Formula 1 manufacturers also competing for the job. Not bad, I thought, for a guy who had more than a fulltime job helping fill pages for eight or more magazines—each month!

MuscleCarB

In the late 1990s I was also involved in some similar work on six cylinder Aston Martin race engines. This involved a study of mixture quality from the carb to the cylinder side of the intake valve, to ascertain how the fuel mixture behaved. Achieving certain goals here netted 20-plus hp.

About 1985, the late Ken Sperling (of Air Flow Research) got into wet-flow testing big time. The subject of his research was Chevy heads for NASCAR use. Although I did not know Ken that well at the time, we had a mutual friend, who was none other than airflow fanatic Roger “Dr. Air” Helgesen. One day, after a big wet-flow test session with Ken, he turned up at my California shop with a car half full of gear to do wet-flow testing and, he was eager to get started. This, along with some of the late (and corny as it sounds, great) Smokey Yunick’s preaching was my baptism into the world of how fuel/air mixtures react in the induction tract of a single 4-barrel-carbed, high-performance V-8 engine.

Since the Ken Sperling wet flow test era of the mid and late 1980s, the subject of wet flow, for all practical purposes, seemed to be dead. That changed when the late Joe Mondello, with a heap of help from some of his high-tech cronies, introduced a new and innovative wet-flow bench, which hit the speed headlines in 2002. At that point, the whole wet-flow debate re-ignited. But with Joe Mondello involved, what else could we have expected?

Since then, I’ve learned a lot and one thing is for sure: Whatever you think might happen, probably doesn’t happen; and what you never suspected, probably does. Actually, performing wet-flow tests is not that difficult, but it can be time consuming. All you have to do is pass a liquid/air charge through the induction system. The easiest way to do that is to blow it through, but it can also be done by suction.

Although the big Dart wet-flow bench shown in Figure 4.3 measures CFM while wet-flow testing, measuring flow at the same time is not really necessary. When wet-flow testing, what you are looking for is how the mixture reacts in terms of staying atomized, and how it may form unwanted fuel rivulets. Still, wet-flow testing done without accounting for what is going on in the real intake tract is much less effective in power development than it could be. But for many this is a new science. So next I offer insights I have gleaned over a long time span.

Six Wet-Flow  Mistakes to Avoid

It is very easy for a novice in this area to make some fundamental mistakes and, as a result, produce data that is not as applicable to a real life induction system as you may assume. Most of the mistakes actually make what you are working with look worse than it really is; so take note as I run through the most common mistakes made.


Fig. 4.4. Although wet flow could have been better yet, attention to it allowed me to build this street-drivable 383 small block Chevy, which produced just 1 hp shy of 610, along with more than 530 ft-lbs of torque.

Fig. 4.4. Although wet flow could have been better yet, attention to it allowed me to build this street-drivable 383 small block Chevy, which produced just 1 hp shy of 610, along with more than 530 ft-lbs of torque.

 

Fig. 4.5. This shot shows the effort that GM put into the development of the 427 Corvette engine’s cylinder heads. The arrow indicates the fuel ramp, incorporated with apparently little flow loss.

Fig. 4.5. This shot shows the effort that GM put into the development of the 427 Corvette engine’s cylinder heads. The arrow indicates the fuel ramp, incorporated with apparently little flow loss.

 

Fig. 4.6. Looking down the Ultra Pro 427 Corvette intake port clearly shows the fuel shear ramp angling off to the left of the valve guide.

Fig. 4.6. Looking down the Ultra Pro 427 Corvette intake port clearly shows the fuel shear ramp angling off to the left of the valve guide.

 

Fig. 4.7. The way flow numbers increase over stock at higher lifts clearly demonstrates the need for high valve lifts. However, the small improvements made in swirl and wet-flow properties also played a part. With a net valve lift of 0.670 inch, the power figures were measurably up, as shown here.

Fig. 4.7. The way flow numbers increase over stock at higher lifts clearly demonstrates the need for high valve lifts. However, the small improvements made in swirl and wet-flow properties also played a part. With a net valve lift of 0.670 inch, the power figures were measurably up, as shown here.

 

8-A

 

Fig. 4.8. Changes to the chamber of the Ultra Pro–modified 427 Corvette head were minimal. In this instance the really hard work was in the form of the fuel shear ramp. Getting this to work and improving flow was not an easy task.

Fig. 4.8. Changes to the chamber of the Ultra Pro–modified 427 Corvette head were minimal. In this instance the really hard work was in the form of the fuel shear ramp. Getting this to work and improving flow was not an easy task.

Mistake Number-1:

Less-than Optimal Test-Pressure Differentials A lot of wet-flow testing is done with a fixed test-pressure differential of between 28 and 50 inches H2O. Since our main focus here is a racing engine, what is needed is to emulate the depression curve that occurs in a running engine. This means drawing about 100 to 130 inches of depression at low lift (0.050 to 0.100 inch) and about 15 inches at high lift. A fuel wetting problem that exists at even 50 inches of depression at low lift can completely vanish at 120 inches. A problem that does not exist on the bench at 50 inches and high lift can easily show at 15 inches.

Mistake Number-2:

Wrong  Liquid-to-Air Ratio If the excess liquid is put through the system, the wet-flow characteristics are bad no matter what you do. The key here is to take into account that about 15 to 20 percent of the liquid fuel vaporizes by the time it passes through the intake valve. The best fluid/air ratio to use to get meaningful results is a mixture (by volume) that represents a fuel/air ratio of about 15 to 20 percent lean.

Mistake Number-3:

Wrong Surface Tension Water has a much higher surface tension as well as a density some 30 percent more than gasoline. These two factors can upset the spread and shear patterns. Adding alcohol and small amounts of detergent can cut the surface tension to the point where water-based test fluids more closely emulate gasoline.

Mistake Number-4:

Wrong Temperature Running the test fluid and the intake valve at a hot temperature made a difference in the way the streams and shear areas reacted. This area might need more research to get a better correlation but, as of now, I recommend a fluid temperature of at least 150 degrees F.

 Mistake Number-5:

Incorrect Interpretation of Results I don’t quite know how to address this; there are so many aspects to consider. Many wet-flow tests reveal problems that are self evident, but some issues are regarded as a problem when, in fact, they are an asset. Fuel shearing off chamber edges and into vortices is often okay, but when they migrate to the cylinder-wall side of the widest part of the quench pad, it is not okay. As the combustion cycle takes place, raw fuel can end up in the ring land volume and contribute absolutely nothing to the game. The obvious problems are just that—obvious. It’s the more subtle issues that require an experienced eye and that just takes time.

Mistake Number-6:

Incorrect  Flow Model Setup Guess what? The intake port on those heads is fed via an intake manifold and in turn by a carburetor. Each modifies the characteristics of the air/ fuel mix flowing out of the base of the carb. Sure, we can start any tests with just a bare head and valves. But from there, the quest for better wet flow needs to have an intake introduced and then a carb. As for the carb itself, it should have the same boosters as used by the end product, if you want to get real fussy about results.

Wet-flow testing something like a set of big side-draft carbs, such as Webers or the like, with each barrel on an independent runner, presents no great test traps. This, however, is far from the case with a plenum style manifold, such as a single-plane intake for a V-8.

Fig. 4.9. Here is what moderate swirl and attention to wet and dry flow can do for an already well-developed cylinder head. Gains start at 3,500 rpm and amount to 20 ft-lbs and 25 hp.

Fig. 4.9. Here is what moderate swirl and attention to wet and dry flow can do for an already well-developed cylinder head. Gains start at 3,500 rpm and amount to 20 ft-lbs and 25 hp.

 

Fig. 4.10. Here, the approach angle of the manifold runner does not match the intake port, so the fuel tends to target the short-side turn more than we want to see.

Fig. 4.10. Here, the approach angle of the manifold runner does not match the intake port, so the fuel tends to target the short-side turn more than we want to see.

 

Fig. 4.11. Dart’s Tony McAfee doing the initial wet-flow tests on a 23-degree small-block Chevy head. For these tests, he is using a straight stack on the intake. After getting the port close, in terms of fuel distribution within the port and reducing fuel rivulets to a minimum, tests with a representative intake need to be done.

Fig. 4.11. Dart’s Tony McAfee doing the initial wet-flow tests on a 23-degree small-block Chevy head. For these tests, he is using a straight stack on the intake. After getting the port close, in terms of fuel distribution within the port and reducing fuel rivulets to a minimum, tests with a representative intake need to be done.

 

Fig. 4.12. Having a more direct line of approach to the port in the cylinder head from the manifold runner almost always helps sort out and improve wet flow. This 18-degree-style small-block Chevy head has manifold runners that, from this view, align directly with the ports. This is not so on many heads, such as the stock-style 23-degree heads.

Fig. 4.12. Having a more direct line of approach to the port in the cylinder head from the manifold runner almost always helps sort out and improve wet flow. This 18-degree-style small-block Chevy head has manifold runners that, from this view, align directly with the ports. This is not so on many heads, such as the stock-style 23-degree heads.

 

Fig. 4.13. A plane view of a single 4-barrel intake manifold reveals the fact that the ports are either turning with the swirl pattern or against it. On these Chevy heads, the outer manifold runners tend to favor the rotational direction of the swirl whereas the inner runners do the reverse. This affects the pattern of the wet flow as it approaches the cylinder head port. Under ideal circumstances, this difference in flow patterns is set up during the wet-flow development of the manifold/head combination.

Fig. 4.13. A plane view of a single 4-barrel intake manifold reveals the fact that the ports are either turning with the swirl pattern or against it. On these Chevy heads, the outer manifold runners tend to favor the rotational direction of the swirl whereas the inner runners do the reverse. This affects the pattern of the wet flow as it approaches the cylinder head port. Under ideal circumstances, this difference in flow patterns is set up during the wet-flow development of the manifold/head combination.

 

Fig. 4.14. This interesting shot shows the initial rush of fuel just as the intake valve (arrow) opens. Because of the high depression caused by a strong exhaust-chamber-scavenging pressure wave, the fuel shearing at this low lift (about 0.1 inch) is very strong even with a radius approach to the seat itself.

Fig. 4.14. This interesting shot shows the initial rush of fuel just as the intake valve (arrow) opens. Because of the high depression caused by a strong exhaust-chamber-scavenging pressure wave, the fuel shearing at this low lift (about 0.1 inch) is very strong even with a radius approach to the seat itself.

 

Fig. 4.15. Top to bottom, we have the wet-flow pattern of a big-block Chevy at 0.300, 0.500, and 0.700 inch of lift. Note how the “wettest” part of the stream (in blue) becomes greater and spreads wider down the cylinder wall. Any wet fuel still on the cylinder wall on the compression stroke finds its way directly into the top ring land crevice.

Fig. 4.15. Top to bottom, we have the wet-flow pattern of a big-block Chevy at 0.300, 0.500, and 0.700 inch of lift. Note how the “wettest” part of the stream (in blue) becomes greater and spreads wider down the cylinder wall. Any wet fuel still on the cylinder wall on the compression stroke finds its way directly into the top ring land crevice.

It may seem easy enough to just flow one cylinder through the carb and intake and on through  the cylinder head itself. This is often how manifolds are tested. Unfortunately, it’s the wrong way to do it. On a single-plane intake, when the piston of the test cylinder is near maximum velocity and the valve wide open, the valve on the next cylinder to draw is as much as a 0.250 inch off the seat and (assuming a race engine) drawing air at a rate of about 20 to 25 percent of the test cylinder. If that cylinder is not hooked up to a depression that simulates this two-runner flow system, the direction and pattern of any wet flow in the intake does not simulate that of a running engine.

On intakes that utilize a plenum, I usually see some basic mistakes made. For a V-8 equipped with a tunnel ram 2 x 4 intake, don’t forget that at any particular moment there are two valves open to that plenum and all the barrels of the carb(s). That means the wet-flow tests must have air drawn through the induction system, with valves open as in a running engine. At this stage, there is one valve open to the lift value at peak piston speed, which occurs about 70 to 75 degrees after TDC on the induction stroke, and one valve in the cylinder about to draw at a lift equal to 20 percent of the flow of the fully opened valve. This compensates for the test pressure differences that are normally experienced at each valve during a running cycle of each cylinder. Going this route means we don’t need to be concerned with the differing test pressures that are normally seen at each valve. I can’t say that what I have outlined here is optimal, but it is a good starting point. For a depression on the nearly fully opened intake, I use 15 to 20 inches.

MuscleCarB

If I am checking the wet flow of a valve in the low-lift position, I use as much depression as the bench can provide, right up to about 140 inches. Making sure the wet-flow properties are good at the point of valve opening and on up through the low-lift range is important because bad wet flow at low lift encourages bad wet flow at higher lifts.

At first sight, it may seem that going to all this trouble is very important for something like a Pro Stock intake. For a single 4-barrel, it may seem like overkill, but in this case it is even more important to use this multi-open-valve technique to get optimal results.

When physically modeling the system, the rule is “Be sure it flows the way it goes!”

Only if you are investigating chamber wet flow should you deal with only one valve at a time. And doing so means using the right test pressure. Fail here and your wet-flow tests lose half their value.

Written by David Vizard and Posted with Permission of CarTechBooks

sa215GET A DEAL ON THIS BOOK!

If you liked this article you will LOVE the full book. Click the button below and we will send you an exclusive deal on this book. Deal-Button