ws6transam.org AFR195 head testing

February 2003:

The Minirammed 385 project received a large boost in December: - Acquisiton of the cylinder heads! With a target compression ratio of 10.6:1, I needed a head with 68cc chambers to match the Lunati flat top pistons that are installed 0.005" down in the short block. Coupled with a 0.039 inch head gasket, the quench area will have 0.044" of piston to head clearance at top-dead-center. In addition, since this is an internally balanced forged and balanced stroker motor with a three-inch intake runner in the TPIS Miniram, I wanted to have some good flowing ports with decent volume. However, it is still a street motor, so I kept the intake runner volume below 200 cc. Hence, the decision to buy the AFR195 head. Now, on to business. The heads flow really well out of the box. AFR published flow data was within ten cfm at all measured valve lifts. However, I've noted a few details that I think will enable me to exceed advertised flow numbers. I've been working this January on trying to eke some more air flow from these heads. As I complete each step, I've been taking them to the shop for flow testing. The goal is to achieve at least a 75% exhaust to intake flow ratio. I've done six flow tests so far, and exceeded advertised flow numbers on the exhaust port. I have now turned my attention to the intake port, were I am now within a few cfm of my goal. Exhaust versus Intake flow ratio is now well above 75 percent at all lifts. So, take a look at my work, and see if you can apply it to your own heads!

The heads were ordered through Dave at Thunder Racing. There are a couple of other discount distributors that also offer a great price on these heads, with savings of about $50 to 100. Dave upsold me to a set of Titanium retainers because of my intended operation of 6000+RPM on a road race track. Expect a six week lead time from AFR for your heads. Contact me if you are actively shopping for these heads.

Flow Test data is here. The initial test of the cylinder head ports showed that AFR indeed advertises their stock flow potential. The advertised flow versus the measured flow were within six percent. However, a hot rodder is never satisfied with out of the box performance. We compulsively tinker!

Hence, in an effort to improve air flow, I added a 30 degree back cut to the intake valve. Although this made a small (but measureable) improvement at low lifts, it probably if not enough to warrant the cost on this configuration of head and valve. The valves are stock AFR stainless pieces, manufactured in India by a company called "Karvalve". (See air flow test #2)

Here is the exhaust port. It flows pretty good, but I noticed a few features that are artifacts of the CNC program. I assume that because AFR must strike a balance between perfection and economic machining, there is a compromise to the contour around the valve guide. It's quite flat and featureless. Therefore I am going to reshape the port around the guide to make a nice aerodynamic taper. Hopefully I'll pick up a few cfm.

Here's the intake port. Notice the three prominent ridges. These are also (I assume) artifacts of the CNC machine programming. It's a digitally controlled robot and some changes to the angle of attack are necessary to keep the tool fixturing from banging into the combustion chamber. These angle changes result in the big ridges you see. I'll be cleaning them up, along with refining the contour of the aero bump behind the valve guide.

Here's the same intake port after cleanup. I've removed the three large ridges and slightly recontoured the radius around the valve guide. I have not drastically changed the shape of the bowl so I'm not expecting too much of a change. I might put some epoxy into the as-cast void behind the airfoil but I'm going to consult a few experts beforehand.

Here's the exhaust port. I've done a fair amount of recontouring of the long-side radius, ecpecially around the valve guide and directly behind it. It looked like the long-side radius was a little sharp around the guide so I blended it and created a little airfoil behind the guide.

Here's a before & after shot of the exhaust port as seen from the port outlet. Can you see the recontour around the guide and directly behind it? I thought it would make a marked improvement to the flow, but testing showed otherwise: Only a couple cfm improvement at very high valve lifts. (See Table, Flow test #3).

Since the recontour of the roof around the valve guide didn't make much improvement, I put in a call to Scott at Air Flow Research to report my findings. I wanted to know what areas of the port I should be concentrating on to maximize flow potential. His suggestion was to next look at the "D" shape of the port and square it up. So, I grabbed my exhaust gasket and did a trace onto the port outlet. However, in retrospect I decided NOT to make the change until after I try working on the valve seat.

Instead of making the change to the D-port, I looked at the valve seat for some air flow. There seemed to be a small step just before the seat and directly underneath. I attacked these areas with a sanding roll, paying careful attention to the valve seat itself. I was dangerously close to the seat. In the end, I just decided to not worry about the seat and have them recut on Dave's Serdi valve machine, pictured below. The stock valve seats in these heads didn't impress me anyway. When shining a blue high-intensity LED up around the valve head, I could see it from the underside of the port, even after pressing the valve down into the seat. Maybe they are okay, maybe they are not. I'm not gambling, and I'm going to have them redone after I'm done porting. I also took the time to have a back-cut added to the exhaust valve. Well, the work paid off: A net increase of 4 cfm across the board, from .300 lift on up. This is documented on the air flow data page as exhaust port flow test #4. 187 cfm at .500" lift.

After working the seat area of the exhaust, it was time to try recontour of the exhaust port exit. The "D" port is machined into the AFR195's so that they can be used with the maximum number of header designs. It is a feature that also aids in reducing reversion into the port. However, reversion is probably not going to be a problem with my exhaust system, so it made sense to remove the feature. AFR was right. 7 to 11 cfm improvement at .400 inches of valve lift on-up.

One note of caution: There is a water passage between these two ports and the wall thickness is less than .080". Therefore there is no room for additional grinding on the wall. Keep the contouring to a minimum except at the corners.

Big gains above .300" in air flow

I also tried to work over the intake port some more, especially on the short side radius. However, I ran into problems with my newly purchased long mandrel. I discovered that my die grinder was just too fast, and the mandrel instantly bent itself into a pretzel. Well, time to switch tactics. Instead of working the short side radius anymore, I worked instead around the intake valve seat to encourage some better low-lift air flow. Just some light sanding with a 120 grit, then a 240 grit sanding roll to knock down any sharp edges that I could find. The result was a few cfm improvement at all lifts. I think I'll now try working that short-side radius some more.

Results of Flow Test #6 here.

I must put in a word of thanks to Mr. Dave Wood of Wood & Sons, in Williamston, Michigan. Dave has been assisting me with the development of this engine in many ways. He has graciously helped me in the flow testing of these heads.

Dave Wood specializes in high performance motors for the Mid-Michigan racing community. He is a craftsman who pays close attention to detail. If you need it done right, call him at (517)655-2998.

Last modified on February 6, 2003

	The heads were ordered through Dave at Thunder Racing. There are a couple of other discount distributors that also offer a great price on these heads, with savings of about $50 to 100. Dave upsold me to a set of Titanium retainers because of my intended operation of 6000+RPM on a road race track. Expect a six week lead time from AFR for your heads. Contact me if you are actively shopping for these heads.
Flow Test data is here.	The initial test of the cylinder head ports showed that AFR indeed advertises their stock flow potential. The advertised flow versus the measured flow were within six percent. However, a hot rodder is never satisfied with out of the box performance. We compulsively tinker!
	Hence, in an effort to improve air flow, I added a 30 degree back cut to the intake valve. Although this made a small (but measureable) improvement at low lifts, it probably if not enough to warrant the cost on this configuration of head and valve. The valves are stock AFR stainless pieces, manufactured in India by a company called "Karvalve". (See air flow test #2)
	Here is the exhaust port. It flows pretty good, but I noticed a few features that are artifacts of the CNC program. I assume that because AFR must strike a balance between perfection and economic machining, there is a compromise to the contour around the valve guide. It's quite flat and featureless. Therefore I am going to reshape the port around the guide to make a nice aerodynamic taper. Hopefully I'll pick up a few cfm.
	Here's the intake port. Notice the three prominent ridges. These are also (I assume) artifacts of the CNC machine programming. It's a digitally controlled robot and some changes to the angle of attack are necessary to keep the tool fixturing from banging into the combustion chamber. These angle changes result in the big ridges you see. I'll be cleaning them up, along with refining the contour of the aero bump behind the valve guide.
	Here's the same intake port after cleanup. I've removed the three large ridges and slightly recontoured the radius around the valve guide. I have not drastically changed the shape of the bowl so I'm not expecting too much of a change. I might put some epoxy into the as-cast void behind the airfoil but I'm going to consult a few experts beforehand.
	Here's the exhaust port. I've done a fair amount of recontouring of the long-side radius, ecpecially around the valve guide and directly behind it. It looked like the long-side radius was a little sharp around the guide so I blended it and created a little airfoil behind the guide.
Here's a before & after shot of the exhaust port as seen from the port outlet. Can you see the recontour around the guide and directly behind it? I thought it would make a marked improvement to the flow, but testing showed otherwise: Only a couple cfm improvement at very high valve lifts. (See Table, Flow test #3).
	Since the recontour of the roof around the valve guide didn't make much improvement, I put in a call to Scott at Air Flow Research to report my findings. I wanted to know what areas of the port I should be concentrating on to maximize flow potential. His suggestion was to next look at the "D" shape of the port and square it up. So, I grabbed my exhaust gasket and did a trace onto the port outlet. However, in retrospect I decided NOT to make the change until after I try working on the valve seat.
	Instead of making the change to the D-port, I looked at the valve seat for some air flow. There seemed to be a small step just before the seat and directly underneath. I attacked these areas with a sanding roll, paying careful attention to the valve seat itself. I was dangerously close to the seat. In the end, I just decided to not worry about the seat and have them recut on Dave's Serdi valve machine, pictured below. The stock valve seats in these heads didn't impress me anyway. When shining a blue high-intensity LED up around the valve head, I could see it from the underside of the port, even after pressing the valve down into the seat. Maybe they are okay, maybe they are not. I'm not gambling, and I'm going to have them redone after I'm done porting. I also took the time to have a back-cut added to the exhaust valve. Well, the work paid off: A net increase of 4 cfm across the board, from .300 lift on up. This is documented on the air flow data page as exhaust port flow test #4. 187 cfm at .500" lift.
After working the seat area of the exhaust, it was time to try recontour of the exhaust port exit. The "D" port is machined into the AFR195's so that they can be used with the maximum number of header designs. It is a feature that also aids in reducing reversion into the port. However, reversion is probably not going to be a problem with my exhaust system, so it made sense to remove the feature. AFR was right. 7 to 11 cfm improvement at .400 inches of valve lift on-up. One note of caution: There is a water passage between these two ports and the wall thickness is less than .080". Therefore there is no room for additional grinding on the wall. Keep the contouring to a minimum except at the corners.	Big gains above .300" in air flow
I also tried to work over the intake port some more, especially on the short side radius. However, I ran into problems with my newly purchased long mandrel. I discovered that my die grinder was just too fast, and the mandrel instantly bent itself into a pretzel. Well, time to switch tactics. Instead of working the short side radius anymore, I worked instead around the intake valve seat to encourage some better low-lift air flow. Just some light sanding with a 120 grit, then a 240 grit sanding roll to knock down any sharp edges that I could find. The result was a few cfm improvement at all lifts. I think I'll now try working that short-side radius some more.	Results of Flow Test #6 here.
I must put in a word of thanks to Mr. Dave Wood of Wood & Sons, in Williamston, Michigan. Dave has been assisting me with the development of this engine in many ways. He has graciously helped me in the flow testing of these heads.	Dave Wood specializes in high performance motors for the Mid-Michigan racing community. He is a craftsman who pays close attention to detail. If you need it done right, call him at (517)655-2998.