MAKING HEADS OR TAILS OF DETONATION "COMBUSTION SCIENCE AND THEORY"

The "internal combustion engine" is still being improved and new understandings seem to happen daily. The key to why we are still trying to figure the thing out is in the name, "Internal Combustion." You can't see it. Yes, there are ruby cylinders and windows that give us clues, but most instrumentation falls short when dealing with 9000-RPM 3Hp/per cubic inch detonating engines. In practical terms a theory is a useful way to account for what is going on when you don't have all the facts.
The following list of facts when put together make a theory that will hopefully enlighten some and will make some sense of the combustion puzzle.
1. I can melt the top of a KB Hypereutectic Piston in 60 seconds with an oxy-acetylene rosebud torch (6300°f).
2. I can melt the same piston in an engine in 3½ seconds with possibly a 4800°f peak temperature during one of four cycles. The intake, compression, and exhaust cycles should be cool down cycles by comparison.
3. When detonation occurs cylinder pressure can see a 500% increase to 5000 PSI.
4. A 2000°f propane torch flame will melt a brick size block of solder given sufficient time.
5. A big piece of copper rod at 1000°f will melt the solder block upon contact.
6. A flame impinging on a cold solid object goes out before making contact.
7. The same flame in a pressure chamber burns closer to the solid object before going out.

We hope to find a theory to explain why detonation can be so devastating and quick. Number one fact indicates to 6300°f, while very hot, does not instantly melt the piston. Number two fact indicates that chamber temperature is less than an acetylene torch and is intermittent to boot. It suggests that something else is responsible for a meltdown engine. Number three, 6 and 7 suggest that pressure may be a player in explaining why a piston can melt so quickly. Facts 4 and 5 show materials melt quickly if you can elevate their surface temperature directly. Back to number 7, increased pressure brings the flame closer to your relatively cool pistons at 5000-PSI detonation conditions -- it must be real close. In fact, my theory is that heat from combustion at 5000-PSI soaks into the piston much as a hot copper rod does solder. Does the flame really touch the metal? Wouldn't that mean the skin temperature a couple of atoms deep could approach combustion chamber temperature and, if so, that hot skin could transfer heat through the aluminum by direct conduction. A somewhat related phenomenon is a cutting torch. A big chunk of steel takes forever to get hot enough to cut with oxygen, but get a little puddle started and you don't even need acetylene anymore. The oxygen burns the steel, but why so fast? Most of the steel is not burned - most is melted while in contact with solid cool steel. No flame available can melt steel at the rate that burning steel can melt steel. Why? No air gap. Heat generated in the burning molten steel puddle is delivered to the cold steel via direct conduction. A large laser is an intense radiant energy source that can burn quick. Again, there is no "air gap insulation factor" and the energy goes directly to the target.
Whether you agree with my theory or not, the real point here is that detonation is the ultimate limiter of Hp in spark ignition internal combustion engines. We use race gas, alcohol, intercoolers, and special combustion chamber designs all to push the detonation limits. We have made eight improvements on KB Piston design recently and all are designed to improve the combustion process - some rescue the area of the piston seeing heat and reduce hot spots. Some of our Step Design pistons re-atomize wet flow and promote quicker combustion. Even if we develop the best piston in the world, if things are going to work, you must do your part -- build fast burn engines. If the combustion process is quick there will be less combustion variation from cylinder to cylinder and cycle to cyle. Normally aspirated engines need maximum cylinder pressure to occur about 12° after TDC in all cylinders every cycle (supercharging could delay maximum desired cylinder pressure a few degrees). To achieve maximum cylinder pressure at the proper time every time, minimum spark advance is what we need to strive for. After all, spark advance does create negative torque, so less is better as long as we can still hit our target maximum cylinder pressure event.
A single side tight quench is a very important combustion enhancer that an engine builder can use to speed flame travel. What quench does that doesn't happen with high swirl or tumble chambers is TDC drive. A normal ignition system starts a small flame about 30° before TDC that burns relatively slow. The piston stops just short of hitting the cylinder head, a blast of air is forced from the quench area through the previously lit flame front. The flame front is driven through the chamber, increasing the chamber burn rate sufficiently to build maximum cylinder pressure early after TDC. The old way to get maximum cylinder pressure early was simply to run 40° of timing. The problem with the old way is that a long slow burn is more affected by hot spots, fuel distribution, spark scatter, chamber temperature, and general cylinder-to-cylinder variation. Don't forget the negative torque component of 40°.
Efficient engines must receive maximum power from all cylinders every cycle. The "all cylinders every cycle" is the hard part. Some of the biggest improvements in smog, power, and economy have come from our improved ability to make all combustion chambers burn at the same rate cylinder to cylinder and cycle to cycle. If you have an engine that #1 cylinder likes 24° timing and #2, 3, and 4 work best with 28-30°, there is a problem. Set ignition at 24° timing and #1 works good and #2, 3, and 4 go along for the ride. Set the timing at 29° and you increase power from the rear three cylinders and #1 goes into self-destruct detonation. Maximum cylinder pressure in #1 may occur at 2° ATDC. At 2° ATDC #1 piston is just getting started down the cylinder. Its slow movement allows cylinder pressure to increase above normal. The increased high temperature, pressure, and time heats the piston and chamber surfaces, encouraging too fast a burn on the next cycle.
Anyone who has run a water-cooled engine without water for an extended time has learned one thing ... an overheated engine pings like a stick on a picket fence. Why? Because chamber heat and hot spots speed up the combustion process to the point of detonation. Fuel burn is so fast that maximum cylinder pressure can occur before the piston reaches TDC. At this point the burning fuel can't expand to move the piston (piston is sitting still) and reduce the chamber temperature. Expansion is a cooling process and without it the fuel burn rate temperature and pressure more or less go critical. This is detonation time. The flame front loses its identity and the entire chamber lights off like a florescent light. Once detonation starts chamber parts become seriously overheated. 600°f chamber parts can ignite the fuel mix making timing or spark retard ineffective in limiting detonation. Ideally the ignition system should be able to anticipate fuel burn rate and adjust the ignition timing to avoid detonation before it occurs. Easier said than done. There is a multitude of variables that affect burn rate directly and indirectly. The most challenging is combustion chamber heat because of the many variables affecting it. If you increase the Hp of an engine 50%, you are going to have approimately 50% more waste heat to get rid of. Combustion chamber temperatures are easier to control in water-cooled engines than air-cooled. Both have one problem in common - the piston is only cooled indirectly. It is difficult to water cool pistons, but it can be done with water injection into the intake system. In my opinion, a "M/L Research" (Ph. 352-357-1005) oil spray system on the underside of the pistons is more practical and does not affect the fuel system. If you study the head fins on air-cooled engines it will be noticed that most of the fin area is on the head with much less cooling to the cylinders and block. Depending upon design, the piston can be exposed to a very similar amount of heat energy that the cylinder head sees. There is some piston cooling through contact with the cylinder walls. At wide-open throttle the incoming fuel charge cools the piston and even wasteful wet flow helps cool the piston. As power levels rise (NOS, turbocharging, supercharging, high compression ratios, and RPM) the danger from overheated pistons becomes critical. Most all turbo diesel and many successful NOS engines are now using oil spray piston cooling and engine oil coolers. Chamber and piston cooling is not an end in itself, but chamber temperature variation does change spark timing requirements dramatically.
Our line of forged pistons is very thermally conductive, and in some cases this will help keep you out of detonation. There will be more heat going into a forging and the diameters will grow more than a hypereutectic, but it will be slightly easier to keep the forgings cool. The recent developments in the hypereutectic KB's are being applied to the forging line. Check for details in the new piston design article. To date, it appears that our special design detonation mini-grooves have reduced the broken top land failure mode, even in full detonation engines. Once detonation raises the piston temperature above what we like, the piston expands to contact the cylinder wall and is cooled, often shutting down detonation before serious damage can occur.
To design and build a fast burn engine make sure all cyliners receive the same fuel mix, compression pressure, timing direction of flow into the cylinder, cam timing, cooling, quench distance, intake exhaust flow, and a similar amout of wet flow and fuel mix cooling. This is a tall order and direction of flow deserves more explanation. Have air fuel flow approach all head ports the same. Most FI systems do this with special manifolding. Manifolding is more difficult with a 4 BBL carb. Typically air flow continues a left turn into the cylinder. #8 cylinder makes a similar left turn to the head but then turns right into the chamber. The change in direction makes a different flow pattern in the cylinder and this changes the burn rate (timing requirement). If our engine has left-right turn and right-left turn due to manifolding and valve layout, the suggestion is to modify your manifold so the first turn is as far away from the head as possible. Even a 95° first turn can help modify the airflow so the flow direction is more similar entering the chamber in all cylinders. I have done some work in this area and recommend it even if you lose some on a flow bench.
Once equal burn rates are achieved timing can be optimized. Do not set timing by 60-foot time or how far wheels will spin from a stoplight. A good indicator of maximum Hp is MPH after 1/4 mile. Even this may put some cylinders into detonation because even our best effort is going to show some cylinder-to-cylinder variation. A maximum Hp setting will likely have four cylinders wanting more timing advance and four cylinders liking a little less. Use coldest plug that works and back up timing a degree or two, and you won't be so hard on the cylinders that like a little less timing. When all is done right you will have cylinders that all like the same ignition timing. When tuned together big Hp, economy, and emission improvements can be had. You may now be able to raise compression ratio and start all over.
Your engine is like a recipe -- add ingredients or change the amount of anything and you have a new dish. Some of our newest releases have valve reliefs designed to re-atomize wet flow in the chamber, for example the KB366. The new Step/Dish pistons reduce hot spots, providing quench and useable compression ratio while driving the flame to the far cylinder wall.
Drive safely!

 

That is a good article.

http://www.kb-silvolite.com/article....n=read&A_id=34