Rattle and Hum
Wayne Tullis; Coquitlam, British Columbia, Canada: I am having issues with detonation in my '71 Nova. I have built a 383 stroker with 10.71:1 compression made up of:
- Forged flat-top pistons with 4cc valve reliefs
- Ported Edelbrock 64cc Performer RPM heads
- 4.030-inch bore and 3.75-inch stroke
- 0.020-inch deck height and 0.041-inch head gasket
- Comp 274 XE with 230/236 degrees duration at .050, 110 lobe separation and 0.490-inch lift
- Comp 1.6:1 roller rockers
- Edelbrock RPM Air Gap Intake
- Holley 750 DP, 81 primary and 84 secondary
- MSD distributor and coil with 18 initial and 34 degrees total at 3,000 rpm
- 94-octane Chevron RON+MON/2 (Canada)
- Five-speed manual and 3.73:1 gears.
The car doesn't overheat and pulls very strong. When I put the engine under load at low rpm, I get detonation. At high rpm, I don't notice it. I have had it on a dyno, and we could not tune out detonation unless we backed the timing way down, which created a loss of power. Air/fuel mixture is good. I have checked the cranking compression and have 205 psi. I have retarded the cam 4 degrees to bleed off a little cylinder pressure. This helped but did not solve the problem. I am thinking of running a 284 XE cam to bleed off a little more cylinder pressure and running a thinner 0.019-inch-thick head gasket to get my quench more in line. If you would suggest the 284 or other cam, would it be with 1.5:1 or the existing 1.6:1 rockers? Is the 205 psi higher than what I can safely use with my fuel or is it another issue? Thank you in advance for any help you can provide.
Jeff Smith: It's interesting how many different questions all seem to eventually lead back to a few common issues. You have also made this a little easier by including excellent information on how your engine was assembled. It is this information and your own suggestion that offers clues as to why your engine detonates. In the engine description, you mention that in the current configuration the pistons are 0.020 inch below the deck height, and you are using a 0.041-inch-thick head gasket. This puts the piston-to-head clearance (also called the quench) at 0.061 inch. This is the distance between the top of the piston and the cylinder head, with the piston at top dead center (TDC). As the distance increases, the compression decreases. The downside is that this additional clearance reduces the quench. The idea behind quench is to squeeze a portion of the air/fuel mixture between the piston and the head so as the piston nears TDC, it pushes the air/fuel mixture into the far portion of the chamber. This increased activity more thoroughly mixes the air/fuel mixture. This generally improves combustion performance and can often result in additional power while simultaneously reducing an engine's sensitivity to both ignition timing and detonation. All of these are positives. We can't overlook the fact that decreasing the piston-to-head clearance (improving quench) also increases the static compression ratio, which might seem counterintuitive, but in most cases, improving or tightening the quench generally improves performance. Your idea to reduce the head-gasket thickness from 0.041- to 0.019-inch thickness is good, as it will improve the quench. But this will raise your compression ratio to 11.29:1, which is a healthy gain of over half a ratio. While this may be a step in the right direction, it is still a pretty serious static compression, and your cranking pressure will increase.
Before you go to this effort, it might be a good idea to try another idea. You mentioned that you are currently using 18 degrees of initial timing with 34 degrees of total timing. One trick that might help is to reduce the initial timing to 16 degrees, and then add two more degrees of mechanical advance. This will create the same 34 degrees of total timing, but it might help with the low-speed detonation problem. Another trick might be to slightly slow the rate of mechanical advance. You didn't mention when the engine rattles, but let's assume it occurs at around 3,000 rpm. Most quick curves have the timing fully advanced by 2,500 rpm. If you slow the rate of advance down by using heavier springs on the mechanical advance, it may delay the onset of detonation at the lower engine speeds. Here's why this might work: All engines create maximum cylinder pressure at or near peak torque. Peak torque is primarily determined (though there are dozens of contributing factors) by cam timing. As you increase cam duration, the engine's peak torque moves up in rpm. That's why going to that larger 282 cam you suggested would also work. The longer duration will raise the rpm point at which peak torque occurs. But before you do this, there are a couple of other less-labor-intensive and less-expensive methods to try. If peak torque occurs around 3,800 rpm, it's possible that delaying maximum timing until after that point might resolve the detonation issue. You can simulate this by just pulling initial timing back by 2 degrees. If that works, you can try delaying total advance. You didn't mention whether the MSD distributor is equipped with vacuum advance. If the engine rattles at part-throttle, it could also be that there is too much vacuum advance at that point. The quick test for that is to disconnect the vacuum advance and see if it still rattles. If not, then you might be able to restrict the vacuum advance to keep the engine out of detonation.
If the previous ideas don't solve the detonation, there's another possibility. MSD offers a digitally programmable MSD 6AL-2 CD ignition box (PN 6530; $359.95, Summit Racing), which offers the opportunity to use MSD's Pro-Data+ software to create a custom digital ignition curve. Here's a quick overview. You lock out the distributor mechanical advance and use the software to create a digital version. The reason you would want to do this is because the digital version offers a way to create a non-linear curve. Start the procedure by digitally duplicating your 18 degrees of initial timing and your mechanical advance curve. Then test to see at what rpm the engine begins to rattle. Let's assume that the detonation begins at 4,000 rpm and is gone by 4,500. All you have to do is pull the timing back by roughly 4 degrees between 4,000 and 4,500. The rpm band between 4,000 and 4,500 is the only place where you have to retard the timing.
In essence, you have built a timing curve that is a dedicated version of what a detonation sensor would produce. You might even experiment with larger amounts of timing retard, as often an engine will experience trace detonation that you cannot hear but yet can still affect power. The beauty of the digital curve is that it is very easy to modify the curve in as little as one-tenth degree steps in 100-rpm increments, if necessary. Of course, there are other ways to accomplish this same task, with items such as add-on detonation ignition retards from a company called J&S Electronics, which we've previously mentioned in this column. There's also water injection from companies such as Snow Performance, which makes a kit specifically for normally aspirated carbureted engines that can be triggered by using a manifold absolute pressure (MAP) sensor and/or rpm to trim cylinder pressures and prevent detonation (PN 20020MC; $550, Snow Performance).
Autotronic Controls Corp. (MSD)