Even the competition has to agree, Chevy's LS-engine family is more than just a worthy successor to the original small-block; it's one hell of a motor. The Blue Oval boys are jumping up and down about their new 5.0L, but (as usual) they are still behind the eight ball in terms of displacement and power output. While the new four-valve 5.0L mod motor offers reasonable high-rpm power, it is decidedly lacking in low-speed power compared to the LS3. Credit the 6.2 liters of displacement for all that wonderful torque. High rpm power is all well and good, but the vast majority of spirited (street) driving comes lower in the rev range. Besides, in the LS3 there is no choice between low-speed and high-rpm power, as the GM motor offers both. Toss in the fact that the LS3 features lightweight, all-aluminum construction, a composite intake, and even variable cam timing (in the automatic/L99 version) and you have a small-block with all the technology of a DOHC motor, without the penalties in size, weight, and complexity. Back in the original muscle car era, it took a big-block to muster power ratings that exceeded 400 hp and a like amount of torque-using those old (gross) power ratings and not net! The modern LS motor makes this a good time to be a Chevy owner.

The LS engine family has evolved constantly to keep GM ahead of the competition. The original LS1 was a solid step above the LT1, just as the LT1 easily eclipsed the performance of the previous L98 TPI motor. In its latest form, the LS3 has followed the 5.7L LS1/LS6 and 6.0L LS2 engine configurations by starting with an increase in displacement, checking in at 6.2 liters. This came courtesy of an increase in bore from 4.00 inches (in the LS2) to 4.065 inches, though the two shared the same stroke of 3.622 inches. The increase in bore size not only upped the displacement, but also airflow, as head flow increases with bore size. Combine this with a revised cylinder head that replaced the cathedral-port design with a more conventional rectangular port, and you have the makings of one serious small-block. Tested on the flow bench, production LS3 heads flow as much as 315 cfm right out of the box. Not long ago it took a pretty serious 23-degree small-block head to achieve these same flow numbers, offered by the stock LS3. Despite flow figures that suggest supporting over 600 hp (we recently made 690 hp with a set of stock LS3 heads on a 468 stroker for our sister book Hot Rod), additional flow is available with proper porting. What we wanted to find out was just how much power was available with a set of ported LS3 heads.

The massive head flow offered by the stock rectangular-port heads means the LS3 responds favorably to cam swaps (Ed note: "favorably" is an understatement!). The flow potential of the stock heads allowed GM to reach its power goals for the production LS3 using relatively mild cam timing, akin to the early LS6 (and LS2) grind. Using the same 204/211 duration split (measured at .050), the stock LS3 cam offers .551/.522-inch lift with a wide lobe separation angle of 117 degrees. This combination of mild cam timing and sizable head flow means the LS3 motor is just begging for wilder cam timing. Think of the LS3 as an all-aluminum race motor without the race cam. Swapping in the right cam will literally transform your mild-mannered stock motor into a street screamer, but where does that leave ported heads? Will even more head flow improve the power output with the stock cam? Are the gains offered by ported heads greater after you've performed a cam swap? Is porting worth the extra expense? That's what we came to find out.

To illustrate the potential lurking in the LS3, and to answer the cam versus cylinder head question, we embarked on a series of tests using a GMPP LS3 crate motor. Performing the multi-head and cam swap on the engine dyno was much easier than had the LS3 been installed in a '10-or-newer Camaro. To illustrate the gains offered by both head and cam swaps, we ran the GMPP LS3 crate motor on the engine dyno in stock trim, then after the installation of a set of GMPP CNC LS3/L92 heads followed by a Comp LSr hydraulic roller cam. To finish things off, we installed one final set of dedicated LS3 castings from ProComp Motorsports. This way, we compared ported stock castings to a set of (affordably priced) aftermarket LS3 heads. Before we could run the GMPP LS3 on the engine dyno, a few changes were necessary. Off came the factory drive-by-wire throttle body and on went a more dyno-friendly manual throttle body from FAST. The stock exhaust manifolds were also replaced by a set of 1 3/4-inch, long-tube headers. The production water pump was likewise ditched in favor of a Meziere electric unit (the LS3 ran sans accessories).

All of the dyno testing was performed with the stock injectors controlled by a new Holley Dominator EFI engine management system. After filling the crankcase with Lucas 5W-30 (non-synthetic) oil, the motor was treated to a break-in procedure prior to running in anger. After the break-in and some quick tuning by Westech's Ernie Mena (the new Holley system was easy to set up and tune), the LS3 belted out some pretty impressive numbers. The 6.2L offered peak numbers of 493 hp at 6,000 rpm and 485 lb-ft at 4,700 rpm. Torque production exceeded 450 lb-ft from 3,900 rpm to 5,700 rpm, and over 400 lb-ft from 3,000 rpm to 6,300 rpm. These are the kind of numbers we'd expect from a stout 383 stroker small-block equipped with a healthy hydraulic roller cam, aftermarket aluminum heads and a good dual-plane, high-rise intake. Even in stock trim, this LS3 crate motor thumped out some impressive power numbers. The difference between the rated and as-tested power output of the GMPP LS3 crate motor can be attributed to a number of different variables. The factory power rating for an LS3 Camaro comes with all accessories, full air induction and exhaust utilizing the factory tune. We changed all these test parameters in favor of the LS3, so our baseline power output was higher.

First on our test list was a set of CNC-ported heads from GMPP. We were curious to see if the ported heads offered any power on this otherwise stock LS3 motor. Unfortunately, the CNC-ported L92 heads from GMPP featured stock valve springs. While adequate with the stock cam, they would be insufficient for performance use with almost any aftermarket cam, including the over .600-inch lift stick supplied by Comp Cams for this test. In terms of valve springs, GMPP offered a set of LS6 springs as an upgrade for the stock springs supplied with the CNC-ported heads. Unfortunately, they provided little (if any) additional spring pressure over the stock springs. What they did offer was additional lift before coil bind, but even the LS6 springs were deemed insufficient for use with our high-lift Comp Cam. Knowing the high-lift cam was in the cards after this head swap, we replaced the stock valve springs with a set of 918 beehive springs (PN 26918) from Comp Cams. The Comp spring upgrade offered an increase in seat pressure of 30 pounds over the factory springs, while providing adequate coil-bind clearance for the high-lift cam. We also installed a set of E3 spark plugs with Diamond Fire technology. Combined with the factory coil packs, the LS3 ignition never missed a beat.

Given the mild cam timing and already impressive flow figures offered by the stock LS3 heads, we suspected the head upgrade would offer only minor power when combined with the stock cam, and we were right. Running the CNC-ported heads on the otherwise stock LS3 resulted in a jump in peak power from 493 hp to 505 hp. Peak torque was up slightly, from 485 lb-ft to 498 lb-ft, though at a slightly lower (4,600 rpm) engine speed. It is interesting to note that the CNC-ported heads from GMPP improved the power output throughout the tested rev range, from 3,000 rpm to 6,500 rpm. Torque production from the aluminum small-block was nudging 500 lb-ft while the torque curve exceeded 450 lb-ft from 3,700 rpm to 5,800 rpm, making for one sweet (and useful) torque curve. The results of this head test are not surprising, since the stock heads already flow well enough to support well over 600 hp. Given the cost of a set of ported LS3 heads, the additional 10 hp didn't seem like the ideal choice on the stock motor. Maybe we'll have better luck with the cam.

With more than enough head flow at our disposal, we turned our attention to swapping out the cam. On our GMPP crate motor, the factory LS3 cam was secured to the timing gear via a single retaining bolt. Most aftermarket LS cams (including the one tested from Comp Cams) featured the early 3-bolt retaining pattern. It was therefore necessary to replace the factory upper timing gear with a new LS2 version that featured both the 3-bolt pattern and the requisite cam sensor lobes. The cam sensor was located on the back of the cam on early LS motors, but the LS3 featured a cam sensor mounted on the front cover. We purchased our new timing gear directly from a local GM dealer (PN 12586481) for around $30. Compared to an $89 cam special for the original small-block, hydraulic roller cams are not cheap, but the cost is reduced somewhat by reusing the factory hydraulic roller lifters, rockers, and pushrods. Hardened pushrods are a good idea with additional valve spring pressure, but we reused the stock pushrods during our dyno testing.

Performing a cam swap on an LS motor is a breeze compared to a traditional small-block, especially on the engine dyno. There is no need to remove the intake manifold, oil pan, or even lifters. Simply rotate the cam and the plastic retainers hold the lifters in place while you swap the cams. With our new cam in place, some minor tuning (basically more fuel) was in order, and after dialing in the air/fuel and timing curves, we were rewarded with peak numbers of 567 hp at 6,400 rpm and 519 lb-ft at 5,300 rpm. The Comp LSr (281LRHR13) cam upgrade not only improved the peak power numbers significantly, but offered gains through most of the rev range. Big chunks of power came after 4,000 rpm, but even down at 3,000 rpm, the cam offered an additional 3-5 lb-ft of torque. Judging by the results, we'd have to say upgrading the cam is a much better choice than porting the heads. Obviously if you have the money to do both, the combined gains are the way to go, but if you can only afford one-go for the cam.

Despite, having replaced the stock heads and cam on the LS3, we decided there was time for one more test. The reason CNC-ported LS3 heads are so popular is their price. The stock castings were so cheap from GM that the aftermarket was initially reluctant to try to compete. Eventually, aftermarket companies stepped up and designed their own castings. The idea behind most was to reduce the port volume of the factory castings without reducing the flow rate. The theory is that more air through a smaller volume equals increased cylinder filling, which in turn equates to more power. ProComp Motorsports was one such company, offering a fully ported LS3 head using its own casting. The LS3 heads from ProComp Motorsports featured 274cc intake ports, 70cc combustion chambers and a 2.20/1.60 stainless steel valve package. As with the GMPP CNC heads, the ProComp heads were run with the factory 1.7 ratio rockers. On our mild LS3, the ProComp heads offered a tad more peak power than the GMPP heads, upping the peak numbers to 573 hp and 520 lb-ft of torque. Just as the head test with the stock cam illustrated, the gains offered by the ProComp heads will increase on a wilder (and larger) engine combination.

  • «
  • |
  • 1
  • |
  • 2
  • |
  • 3
  • |
  • 4
  • |
  • View Full Article