SBM Edelbrock/Speedmaster Head Data

Here's a couple of pictures of the pushrod pinch on the Speedmaster port that's currently on the bench. I've finally switched to a 2.08 valve after working forever with the 2.02 valve. Here are some opinions from my perspective, and I do not claim to be correct.

Regarding the pushrod pinch, I would take more width over more height any day. The pushrod pinch is not THE problem with these heads, but it is a problem. As PBR often said, the stock location of the pushrod can support about 330 cfm. I have only been able to get to about 311, but I know more is there. THE problem with these heads is a lack of usable width at the apex of the short turn. We need to be able to get as wide an apex as possible to get the velocity down over the short turn. The problem that the pushrod creates is that air flowing on the floor of the port has to navigate the pushrod turn and then stay attached to the pushrod wall on its way to bore center. The pushrod effectively limits the width of the flow zone going over the short turn. Tubing the pushrod hole lets as much width as possible be utilized. Raising the roof does do some good, but it also can create problems with flow separation around the pushrod turn. Expanding too fast in an area of high velocity leads to flow separation.

One more comment about the pushrod pinch. Enlarging the pinch should only be done after the downstream problems are addressed. Enlarging the pinch as a first step can actually hurt flow. If you think in terms of Bernoulli and pressure vs flow and flow separation it makes some sense, but here's another way to look at it. At high lift the short side (or short side radius, SSR) is under stress. It's doing all it can to hold on to flow without losing the flow over the SSR. If you enlarge the pushrod pinch and nothing else, you've just put additional stress (additional pressure and flow) on the SSR. It can't handle the additional stress and flow separates earlier and peak flow actually goes down. There is a balancing act between the pushrod pinch and the SSR. The pushrod pinch can actually be used to 'tune' the SSR. It's a bit of a crutch, but these low port heads need help sometimes.

Attached are a couple of flow curves that represent my best efforts so far with both 45 and 50 deg seats on the same port with the 2.08 valve. I've tested this port many times, made modifications, epoxied the throat and seat back up and cut it again. The short side and chamber are now mostly epoxy. The big challenge for me is to raise point B without lowering point A. I've still not quite figured out how to remove the entire dip, or if it's even that important. It's more of a challenge at this point.

As far as heads for a 408 stock block - I guess it depends on how you intend to use it. Street use and under 6000 rpm, it's hard to beat the Trick Flow's for the price. All out racing is going to require something ported, which could be custom Speedmasters, Edelbrocks, Trick Flows or the Bloomers. Just my opinion.

View attachment 1716330913

View attachment 1716330914

View attachment 1716330915

View attachment 1716330916


Looking at your graph I see the little dip in the 50 that starts about .250 lift. It’s more pronounced with a 55.

I tried for years to understand it, but the last I recall about what we came up with was it was a l/d verses the widths of the cuts under the seat.

I know we found a couple of valve jobs from other people that had narrower cuts under the seat and the dip would go away.

The problem was it would flow more air with the narrow cuts but it always lost power.

I guesstimated that with the narrower cuts the valve job was functioning as a radius and that killed power.

It is there on Chrysler heads, Chevy heads and even a 351 Cleveland I didn’t port but flowed. Some of the Chevy testing was done on AFR 227 raised port heads.

To this day I’ve never seen an explanation for that dip. Or why getting it out of the flow curve hurt power other than my WAG on the valve job acting like a radius.

Edit: bad typing. I meant to say where the curve goes through .250 lift.