7 Liter Corvette Heads - can they be improved upon?

[DavidVizard-GFN]

Can those trick 'Vette heads be improved upon?

By

David Vizard

When Chevrolet introduced the 7 liter Corvette they created a stir that’s for sure. 427 inches is hot news in itself but add to that titanium rods, a dry sump lubrication system and a new design of head that sported not only the biggest intake valve in what we commonly refer to as a small block, but also CNC porting. All this high tech stuff on what is essentially a production head warranted looking into. Here at GFN we managed to get a loaner head to take a look at and flow test. It quickly became evident that GM had done a lot of homework on this one. Titanium intake valves of 2.2 inches diameter are pretty big but each one has one eighth of 7 liters to feed. And the traditional 1.6 inch diameter exhaust valve, which may suffice for a 350, now has to handle the exhaust of 7 liters. What all this means is, big though the valves may be, they are not big in relation to the volume of the cylinder they must service. So if the job is to be done effectively the only other option, if sizes are already at a maximum, is to make the valve/port combination as efficient as possible. The question at this point is how well did GM do?

On the Bench.

The arrow here indicates the intake port ‘dam’ situated on the cylinder wall side of the port roof. As unstreamlined as this may look it’s removal cost airflow numbers.

Upon pulling the intake valves from the guides the first thing that becomes evident is that the intake port is far from conventional in form. On the cylinder wall side of the intake port roof there is a diversionary flow dam and it does not look good for absolute flow numbers. First thoughts on this dam is that it must be there to enhance swirl or to enhance the mixing of air with the fuel from the upstream fuel injector nozzle. This form of dam is something that cropped up in cup car head development about 8 – 10 years ago and subsequently faded from the scene. It’s re-emergence here kind of came as a surprise. A look at the nearby shot of the intake port shows the head was CNC’d with a course feed as this is significantly cheaper to do and has far less impact on the heads flow than might at first be expected. The reality of the situation then was that GM was going for low cost and hp - not looky-looky cosmetics. On the bench the head delivered some impressive figures and right then it became very apparent that any attempt to get more air through these heads was going to be difficult at best. As a first thought it seemed as far as air flow was concerned cutting out that dam would be a good move – time would tell.

MTI Racing [www.mtiracing.com] is very much a full service ‘LS’ engine performance shop. Situated in Marietta, Georgia, this company will be familiar to many hard core LS engine performance enthusiasts. Dedicated to development of performance at a level that can realistically be described as amongst the best the US has to offer this companies cars have, when road tested by some of the biggest names in the magazine business, elicited many great compliments. In a business that is as competitive as the high performance industry that is not an easy thing to do on a regular basis. Any company that does regularly and consistently score high points with the elite of the press has to work hard to get there. That means being a leader rather than a follower. When the 7 liter Corvette hit the scene it’s high tech nature meant that any progress called for even greater efforts on the part of the engine developer. The already strong performance of the stock CNC ported heads was as big a challenge as any.



OK - as interesting a the stock head was we got to tie in, so to speak, on the progress of the MTI Racing head mods to the extent of testing some of the interim experimental ports.

During the first few phases of the development the MTI Racing heads featured a reduction in the size of the intake port dam. Surprisingly this cut flow. With it completely removed to make a more conventional port the flow was down as much as 10 cfm toward the top of the lift range. Clearly this was not the way to go. After testing a lot of small variations it seemed that the best avenue was to go back into the port and see if any refinements could be made while still retaining the dam in the intake port.

On the left we have the new high flow port model ready for digitizing. On the right is the original port. To get from right to left the port roof was raised slightly, the dam made a little more pronounced and the bowl area at the one o’clock position deepened about a mm. The short side turn also came in for a little re-contouring but in all, the amount of material removed to get from stock to modified was very small – in the order of about 4 cc’s.

Like the intake the stock exhaust was really good. But there was a little left to be had as it responded to moves that have proved very effective on cup car cylinder head exhaust ports.

After much testing the results in the following chart were seen.

Unless you are used to looking at flow numbers the significance of the overall figures will not really strike home. What we have here is a production head that with minimal reshaping is delivering results as good as a $10,000 a set Cup Car heads. This is called real progress!

Bearing in mind just how good the stock head was the modified numbers are very creditable especially when the fact that the port volume has barley increased is taken into account. Let’s analyze what we have here in terms of power output.


Power Potential.

At low lift, that’s up to about 0.15 or so of the valves diameter, (in this instance with 2.2 intake and 1.6 exhaust that would be 0.330 for the intake and 0.240 for the exhaust) the flow is almost entirely about the valve seat design. Above 0.25 of the valves diameter or 0.25D as it is called (0.550 for the intake and 0.400 for the exhaust) flow is just about all port design. From the stock numbers it could be seen that GM did an outstanding job on the seat geometry of both the intake and exhaust. After a great deal of effort flow gains were seen from improvements in the port form on both the intake and exhaust. But because virtually all the flow increase was from a superior port design all the increases were seen at high lift. Flow gains at high valve lift are always harder to make use of because it makes far greater demands on valve train dynamics to be able to lift the valve sufficiently high enough to put it into the increase flow zone. Fortunately for us we are dealing with an LS style engine, not the traditional small block Chevy. Here’s why we are fortunate.

Even though the intake here is big at 2.2 inches diameter it is still not big enough for 427 inches. Add to this the fact the engine will easily stretch out to 454 with 500 plus inches on the cards. At 427 cubes the intake valve would, to deliver the same flow area up to the critical 0.25 D point, need to be 2.46 inches diameter. It is clearly well short of this so, for a head with only a 2.2 inch intake to work the valve must be lifted well into a high flow zone as quickly as possible. Fortunately the LS range of engines is well situated to do just that far more effectively then a traditional small block Chevy. The key factor is the LS’s design of roller lifter cam and valve train. When the factory decided to go roller only they were smart enough to design the cam with a much larger base circle than any GM engine that had gone before. The significance of this is that the bigger the base circle is the lower the pressure angle the roller lifters experience for a given lifter acceleration.

Unlike a flat tappet cam a roller cam has a significant side load. The steeper the pressure angle the higher the side load. More lifter acceleration means more side load while a bigger base circle on the cam and a larger roller both reduce the pressure angle on a lifter for a given acceleration.

The higher the pressure angle the greater the side load on the lifter. A sane limit is about 28 degrees. This side load puts a limit on how fast the lifter can be accelerated. If either the lifter roller or the cam’s base circle is increased the pressure angle is reduced thus allowing for an increase in lifter acceleration by employing a steeper lobe on the cam profile until the pressure angle limit is once again reached. The sizing of the LS’s components in this area allow for a far faster lifter rise then the old small block. This means that our 427 has a much increased potential for faster valve opening and, if we add to the fact that it has a light titanium intake valve, it is in an even better position yet to successfully employ very high valve lift numbers. This ultimately means the engine is in a strong position to access the increased flow seen at high lift.

On the left is the finished head. The example here is with an optional Ti exhaust valve for race only use. The Ti intake is stock GM issue. The shot on the right clearly shows the bowl area of both the intake and exhaust. While the exhaust looks about what you would expect of a mega buck Cup Car head the intake, with it’s dam, is somewhat unconventional.

The shot on the left is a view down the intake port and clearly shows the revised form of the dam. On the right is the exhaust port. Other than the fact it is a highly developed form it is entirely conventional.

So how does all this translate to increased output?

Flow figures and swirl coificients are all very well for the flow bench. The real test is how all this works out on the dyno. As it happens all the effort that went into this design of head from MTI Racing pays off to the tune of 31 hp and 22 lbs-ft of torque. All this, as can be seen from the curves shown below comes with no negetive impact on low speed output.

This was all achieved by attention to detail to refine the port designs rather than enlarging them. In this instance port cc's only increased by about 5. There is a lesson to be learned from that. Although 560 RWHP is pretty respectable I would have to say there was a lot more that these heads can deliver given a more radical combination of intake and cam. Then there is extra displacement - they should really pay off in this scenerio.

[jerminator96]

Awesome numbers, I can't wait to see what kind of power that 454 puts down.

[SStrokerAce]

Dave,

I've seen a LOT of these heads, and I'm interested in what you get for results on them. I've seen people make from 540-630rwhp with the heads and it's very impressive what Mike Chapman did with those heads.

I think your right on the money about the exhaust on the heads, but even with a small valve 260+cfm is possible out of them.

Bret

[SStrokerAce]

David,

I'm very interested in the dyno results. Are you going to run the heads, stock and ported on the same motor?

Bret

[Dcmgt]

David - You mention the Ti exhaust valves you put in the LS1 heads were for race only. What data or specific experiences do you have to substantiate the restriction to such usage? An S/N curve would really be ideal to come up with a replacement interval for a specific engine. Some input I've received (incl from, Del West) is that the life of the Ti exhaust valves should be quite good, at least 20k miles, if you have low enough spring force.

I'm restoring an old FE Ford factory race engine that came originally with hollow stem valves, sodium filled in the exhaust. Because the exhausts were known for breaking, I picked up a complete set of Ti valves as a replacement that would most closely represent the originals (solid stainless would be too heavy and require stiffer springs which will just exacerbate parts breakage and wear). My open spring pressure will be about 435 lb and with 11/32" stems I'm hoping these will be good for a reasonable life on the street with occasional track use. I understand the Corvette LS1 valves have a smaller diameter stem, which results in a higher stress level for the similar spring load people seem to run in those engines, but what data is there to substantiate any of these assessments? Mike

[DavidVizard-GFN]

Dcmgt,

It's not so much that Ti valves are unreliable they are just not as long lived as steel valves. Ultimatly it is the heat cycles that lead to failure just as much as total miles. But the real reason for not using them is that, for the most part, your engine does not need them. Although intake valve accelleration is important to output the same cannot be said for the exhaust. This means weight is not so important. But there is still one more issue here. A sodium cooled valve will have less tendancy to bounce than you might think. Because th sodium sloshes up and down it acts like a damper so even though the valve may be about 10 -15 grams heavier it does not act that way in practice. Couple this to a less agressive valve action and you can see that the sodium valve will not be a limiting factor in terms of output but it will be more reliable. I have a Cup
Car (2002 Dodge Intrepid) I have rebuilt the engine so it can survive 24 hour races. It uses Ti intake and sodium cooled steel exhausts.

DV

[Dcmgt]

Thanks for the feedback David. Your response prompts some further questions.

First, who may be able to share the data to quantify the Ti valve life? I'm sure the top race teams, at least in F1, would not risk championships and millions of dollars of sponsorship per race on such a part without having very rigid test data to base a life on. If there is a strength reduction from heat cycling, that must have been determined and taken into account.

Next, I understand a number of production engines now use Ti valves, incl exhaust, such as Ferrari, Honda and Yamaha according to this site: 928 DEVELOPMENTS: ANALYSIS Note that they state that Del West has told that author that their Ti valves are as durable as stainless counterparts and not susceptible to heat cycling, just that the seats and tips may wear more. The tip wear issue is resolved by the valves having hardened steel tip inserts. I'm not as concerned with seat wear as I am with loosing the head and lunching the entire engine.

On the exhaust valve action, optimized cams for the old FE Ford use dual profiles. I understand a big part of that is because the ratio of exhaust to intake flow of the original head is not ideal and a more aggressive exhaust profile is used to compensate for it, as well as some exhaust-specific timing issues. Regardless, I already have the Ti valves and cam and just desire substantiation of a replacement interval or evidence that the life would be totally inadequate to justify scrapping them and spending more $ on another option, which may also impact other components. Any help would be appreciated to solidly resolve this issue.

[MAP]

Greetings David,

I have a question about the modifications to the intake bowl area right next to the valveguide. By looking at this area through the manifold-side of the head (in the "Finished Ports" photo,) it's curious to me that the area to the right of the valveguide (i.e., closer to the bore centerline,) which you said was was deepened about 1mm, seems to form a helical channel with roughly a 45-degree departure from verticality as seen in this photo, from bowl roof, approaching the valve seat. If the local air/fuel flow follows this channel, then it would seem that upon exiting the valve, it would be headed for a nearly right-angle crash with the cylinder bore.

It would also seem to direct flow in a way that would be exactly counter to that needed to enhance swirl.

Now knowing that you would never do such a thing, my question is, how are my intuition and understanding failing me? To clarify the dilemma I have, my opinion would be that to enhance swirl, one would want to create exactly the mirror-image of the bowl as you (and GM) made it, dam and all!

Thank you,
Mark

[MAP]

Greetings David,

I'm puzzled about the valveguide region in the intake port. The article mentions that the area to the right of the valveguide (as it would be seen in the third photo down from the beginning of the article,) was deepened about one mm. In the "Finished Ports" photo, it's striking how this depressed area seems to accentuate a helically-shaped path descending from the roughly the mid-height plane of the bowl, down toward the valveseat. In the view captured in this photo, the helix seems to make about a 45-degree angle of inclination with respect to the bowl centerline.

Assuming that a not insignificant portion of the air/fuel mixture follows this path of descent toward the valve, then after it crosses the valve, it would seem to be headed for a head-on crash with the cylinder bore wall. This action seems counter-intuitive to me, since this would seem to reduce flow, and reduce swirl.

In fact, it would seem that a mirror-image of the bowl shape - dam, helical trough and all - would enhance flow and swirl.

Since I'm just a hobbyist, and since I'm sure that the collective braintrust (between GM and you) that developed this port is unimpeachable, then common sense would tell me that my intuition must be wrong.

Could you please enlighten this apparent dilemma?

Thanks and best regards,
Mark

[rsicard]

David: In the write-up on these heads and associated ports, the base circle for the LSX camshaft was increased. How much larger is the base circle than the Gen 1 SBC camshaft? Are the lifter bores deeper or longer in length than the Gen 1+ lifter bores? Please advise. Thanks.