If you get into flowing cylinder heads, you’ll discover there is more than one way to flow a cylinder head. We’ll use one example to show how numbers can change dramatically based on how the cylinder head is tested. We recently flow-tested a CNC-ported L92 GM Performance Parts cylinder head. Our intake flow numbers were within 2 percent of what GM claims in its catalog, which is good. However, what the catalog doesn’t say (and what is critical to know) is that the exhaust port numbers were achieved without a flow tube. We tested our heads with a flow tube. The GM head flow number at 0.400-inch valve lift was 168 cfm while our head with a 17/8-inch flow tube generated a much greater 192 cfm. This is a 24 cfm or 14 percent higher flow. As an average, we’ve seen around a 5 percent higher number with a flow tube, but the ported L92 head really jumped. Neither number is wrong, but it points out how important it is to know if the flow tube was used when trying to decide which number you will use to determine exhaust port efficiency relative to the intake. All these numbers were generated at 28 inches of test depression.