Build Your First Performance Engine: Block Prep and Math
Don't pay the other guy. We'll show you how to build your first performance engine.
If you've been into cars for a while, you've probably swung a few engines over the fender that were assembled by someone else. That's cool, but it's probably about time to build one yourself. Overall, it's really not that complicated, but there are a few things to watch out for if you want to get it right. Even something as simple as pulling a block from a yard has tricks. You've got to know the math before you start to order parts, and it's also important to communicate with the machine shop if you want to get what you paid for.
There are a couple of differences between a standard rebuild and a performance engine, so we are going to include the research techniques and tools available to get exactly what you want.
We are eventually going to use the engine we build as a test mule for heads and cams, so it's going to need to last a long time and take plenty of beatings. That also means the engine should take anything we can throw at it once it's in a vehicle. The goal here is an engine that anyone can assemble and that makes good power for money well spent.
Just Pull the Thing
It's still better to pull an engine from a car than to buy one anywhere else. The simple reason is cost. Every part that's included with a complete engine is a part you don't have to buy, and vehicles that have been wrecked were running and driving right before they ended up in the yard. During our search, we stumbled across two identical, smashed '74 Chevy Caprices and scored two complete 400-cid engines. We figured that was a sign.
So how do you know if the block is good? We always bring a crank socket and barring tool, a compression gauge, and a dial indicator when we are checking out an engine. Most yards will have a stack of batteries up front for some cranking power. Crank the engine five or six times and take the highest reading with the compression gauge. The cylinders should be within 10 percent of one another. Chances are the heads aren't great castings, so we pop them off and check the cylinder bore for wear and taper. The math for displacement is: displacement = pi / 4 x bore2 x stroke x number of cylinders. From there you can solve for bore (bore = displacement / (pi / 4 x stroke x number of cylinders). In our case it was 4.125 inches. We were lucky that both engines had stock or close-to-stock bores. The machine shop will measure final cylinder runout with an inside micrometer, but in the yard it's OK to just run your finger down to the top of the piston and feel for a ridge. If there is a slight ridge and a stock bore, there is room to work.
The biggest buzzkills at the machine shop are cracks in the block, especially on 400s with steam holes and siamesed (thin) cylinder walls. Little cracks are worth the gamble, but the big ones are death and are evident by a trail of rust down the center of the bore or between the head bolt and the cylinder wall. Rust means water entered the bore somehow. Look for cracks in the lifter valley, as well.
We tossed the heads and trundled one of the engines down to JMS Racing to be cleaned and inspected. The machine shop is going to tell you if the block needs to be overbored and whether the crankshaft can be saved. Seems obvious, but this is going to predict the parts you'll need to get the correct compression ratio, piston bore, and deck height. We try to order the bearings, rings, and gaskets from the machine shop and pick them up when the machine work is done. Saves a lot of time.
We usually start with a compression ratio rule of thumb for building a naturally aspirated engine. For a 350-inch engine with modern aluminum heads and a cam at or bigger than 220 degrees of duration, you can run about a 10.0:1 compression ratio. That assumes the quench area is around 0.040 inch and you will be using 92 octane or better pump gas. The quench area is the space between the flat portion of the piston and the flat portion of the head outside the combustion chamber. As the pistons reach top dead center (TDC), air and fuel are squished toward the combustion chamber, increasing turbulence and reducing detonation. Modern head designs utilize a kidney-shaped combustion chamber to accomplish this.
The cam selection will also reduce an engine's proclivity to detonate. As the cam increases in duration, the closing point of the intake valve is moved later. If you take two cams, one with an intake closing point of 54 degrees after bottom dead center (ABDC) and one with a closing point of 60 degrees ABDC, the camshaft that closes the intake sooner (54 degrees, in this case) will create more cylinder pressure than the later-closing, longer-duration camshaft and will be more prone to detonation given equal static compression ratios. The ideal cranking compression is around 180-200 psi—any more will ping on 92 octane. Also, given the same duration, increasing the LDA decreases the amount of overlap and also closes the intake later in crank degrees, reducing the engine's willingness to detonate and increasing peak power.
So, with theory in hand, we set out to create a short-block with a cam that will not encourage pinging on 92 octane with a 0.040 quench and 10.5:1 compression. Before ordering anything, we called JMS for the details on the engine. They said the crank needed to be turned 10/20, 0.010 inch on the rods and 0.020 inch on the mains, and checked for some runout with an inside micrometer to measure the bore just below the deck, again in the middle of the bore, and finally at the bottom. Machine shops will bore the engine to within 0.005 of the final bore size, then finish the last 0.005 with a hone. Our engine needed to be bored 0.030 to a final size of 4.155 inches. Knowing that, we mathed out the rod-and-piston combination.
Piston manufacturers usually provide a ballpark compression ratio based on cylinder-head combustion chamber sizes. Our cylinder heads are AFR 210s that we bought used with 67cc chambers. A quick search on the internet netted us two different off-the-shelf piston combinations that estimated compression ratios between 10.0:1 and nearly 12.0:1. Compression ratio is determined by the ratio of the entire volume of the cylinder with the piston at bottom dead center (BDC) divided by the volume of the cylinder at TDC. We determined the actual compression ratio by adding the cylinder volume to the chamber volume, then dividing it by the cylinder volume. The cylinder volume is easy to calculate; it's simply the displacement formula without the multiple of eight (pi / 4 (0.7853) x 4.155 x 4.155 x 3.750 = 50.84). This gave us a cylinder volume in cubic inches that can be converted to centimeters by multiplying each measurement by the conversion factor of 2.54 and rerunning the formula (0.7853 x 10.550 x 10.550 x 9.525). That gave us 832.540cc total cylinder volume. To get the chamber volume, we needed to simply add the 67cc chamber volume to the gasket and piston deck height. The gasket was 0.039. Converted to centimeters it was 0.099. Then we converted to volume using the old formula 0.7853 x 10.55 x 10.55 x 0.099 = 8.653. To get the depth of the piston in the cylinder, we needed the block deck height from the machine shop minus the total height of the piston-and-rod combination. The rod we selected is 5.70 inches, half of the stroke is 1.875, and the piston- compression height is 1.425, so the total height is 9.00 inches. We asked the machine shop for 9.005 deck height after machining, which should put our piston 0.005 in the hole. Converted to cubic centimeters, an additional 1.110 cc's were added to the chamber volume. Finally, the pistons had two 8cc valve reliefs.
The final math is cylinder volume plus total chamber volume divided by total chamber volume (832.540 + 92.763 / 92.763 = 9.97:1). This is virtually what we shot for and likely will not detonate. We shall see.
After we figured out the compression ratio, it was time to figure out how much power we would be making. For this we used Comp Cams' DynoSim software. The program is divided into six sections. The first section allowed us to select the engine size with overbore, and it provided the stroke and calculated the total displacement based on the engine make and model.
The next section had a list of common cylinder heads, allowing us to enter the airflow data for each lift point and get the exact cylinder-head airflow into the dyno graph. This is where the Cylinder Head Database on CarCraft.com is helpful. A number of cylinder-head flow tests from the same flow bench are posted on the site to cut and paste. Also, AFR has flow data on its website—just make sure the data is from the most recent version of the head.
We know we made you figure out your compression ratio the old-fashioned way, and you probably know that dyno software has a compression-ratio calculator built in. You still have to know the deck height and the cc's of the head to use it, and our math was within 0.004 of the calculator's, so we know it's right.
The next two sections are a little soft, in our opinion, with a selection of intake designs such as "high-flow single plane," and so on. You can neither see nor control the math in this area. The exhaust section basically assumes you are using good headers.
The cam section is highly detailed and offers a great deal of control. You can go to the Comp Cams website and download a custom profile or type individual cam-timing specs into the Cam Manager. We tried three different cam grinds and posted the results in the Dyno Results sidebar. To verify the DynoSim's accuracy, we are going to run the same cams on the dyno at Westech Performance after the engine is built to tell if the software got it right.