Enlarging PRP doesn't always help

-

Earlie A

Well-Known Member
Joined
Aug 22, 2023
Messages
550
Reaction score
707
Location
TN Hills
Speedmaster SBM head with 2.02 intake valve on the flow bench. Let's assume this head is going on a large stroker motor and we're looking for high flow. I'm trying to find the flow limits of the head and understand why flow stops increasing. Throwing all the cheat tactics I can think of at the head, flow has peaked at 308cfm. Why does it stop there? Where is the choke point? Checking velocity at the pushrod pinch and short turn apex shows some insane velocities. These are shown in the first chart under Test 568. I've convinced myself that increasing the size of the PRP has got to help. Lowering the velocity through the PRP will keep speeds down approaching the apex. Surely enlarging the PRP will also relieve a restriction in the port. If you pinch off a water hose the flow goes down. Relieve the pinch and flow goes up. Doesn't work that way with airflow in a port. As test 569 shows, increasing the size of the PRP does not always help. Flow in Test 569 is lower than flow in Test 568.

Everything from here on out is speculation on my part. I'm thinking out loud and sharing, not teaching.

I've run into this a few times before and it can be quite confusing. The PRP is definitely a restriction in the port. Why does relieving that restriction not increase flow? I think the answer has to do with laminar flow vs turbulent flow. In the water hose, flow is laminar, so reducing the restriction increases the flow. But in the intake port flow is not laminar and changing the PRP affects what is happening down stream. Look at the drawing of flow bench Tests 568 and 569. Both drawings depict 0.650" valve lift. In both tests the inlet of the port has a pressure of 0 inches H20 if read on a gauge. In both tests the pressure in the cylinder is -28 inches H2O. I don't know what the actual apex pressure is, but let's assume in Test 568 that the gauge at the apex reads -3 inches of H2O. That gives us a 3 inch pressure drop through the PRP. In test 569, the PRP has been enlarged and the gauge pressure at the apex would now be higher. Let's say it increased to -2.5 inches of H2O. So now there is less pressure drop through the PRP and a higher pressure at the apex. That should be more pressure available to push air through the port. Flow should go up, but it goes down.

What happens down stream is flow separation. At high valve lifts the flow cannot stay attached to the short side. As flow increases, separation increases and shuts down the available flow area. So what might be happening here is this. Even though flow in 569 is less than 568, the potential for flow is greater. The higher apex pressure in 569 is TRYING to push more air through the port but flow separation decreases the flow area so flow ultimately drops.

So here's my takeaway. In general, for an 18 deg SBM head on a large motor, a larger pushrod pinch should be a good thing. But if there are other problems down stream, enlarging the PRP can expose them and even compound those problems. Pittsburghracer has said many times that the PRP is not the problem on the SM/Edelbrock head. I'm starting to understand what he means.

As mentioned before, this is an opinion and a theory. I welcome others.

IMG_2636.jpg


IMG_2637.jpg


IMG_2635.jpg
 
Last edited:
It’s nice to see that some of the things I’ve mentioned over the years has stuck with some guys. I still hear it but not everyday like I used to that guys mention attacking the “pinch”. The pinch needs maximized to available area but only after the shortside is lowered and widened to gain the needed area
 
It’s nice to see that some of the things I’ve mentioned over the years has stuck with some guys. I still hear it but not everyday like I used to that guys mention attacking the “pinch”. The pinch needs maximized to available area but only after the shortside is lowered and widened to gain the needed area
I don't think you realize how much you have helped. Just now you included the word 'lowered'. I'll be thinking about that for days now. I've tried to widen as much as possible, and lay back a little, but never lower. I've been a bit intimidated by the short side. I don't know why, I've got a large can of epoxy. It's just the closer you get to having a good port, the more the fear of messing it up can paralyze you.
 
I don't think you realize how much you have helped. Just now you included the word 'lowered'. I'll be thinking about that for days now. I've tried to widen as much as possible, and lay back a little, but never lower. I've been a bit intimidated by the short side. I don't know why, I've got a large can of epoxy. It's just the closer you get to having a good port, the more the fear of messing it up can paralyze you.


For sure it needs lowering. Remember the other point I tried getting across. The common wall slants in, the roof slants down, and the floor slopes up. Each of those areas plus the pushrod side wall has to be worked.
 
For sure it needs lowering. Remember the other point I tried getting across. The common wall slants in, the roof slants down, and the floor slopes up. Each of those areas plus the pushrod side wall has to be worked.
That pushrod side is about as tricky as the short side. Too tight a radius and you've lost it.

As far as the roof goes, I've found the cyl center side needs to be lower and straighter than the common wall side. On the cyl center side the air wants to go right to bore center. On the common wall side it will follow the roof more.

Thanks again.
 
That pushrod side is about as tricky as the short side. Too tight a radius and you've lost it.

As far as the roof goes, I've found the cyl center side needs to be lower and straighter than the common wall side. On the cyl center side the air wants to go right to bore center. On the common wall side it will follow the roof more.

Thanks again.

A couple time savers for me is after I set my pinch dimensions (on my head rotator) to give me more room to work the port I move on to the common wall and port to the tube. This all gives me better visual and max room. I then flip the head roof side down to raise the roof before I touch the floor. Through experience and testing I have a snap gauge that I use to set the roof dimensions where I want it. Keep this dimension handy as I use it on every valve size from 2.02-2.08. At this point I can work the floor some but I do my finish floor work after I have the head off my rotator. Take a set of 4-6 inch calipers and with the head on its end measure up through the valve seat and measure from the head deck to the top of the shortside. Keep notes of these measurements as you do your flow tests. I have a different height for each valve size and again have this number written down. I wish I could give you that dimension for the different valves sizes but I have 100’s of hours of porting and testing to get that number. My set I’m working for that hopefully will soon get on my duster are flowing 334cfm so far and have a totally different dimension. I didn’t reread this as I have some running to do so I hope it makes sense.
 
A couple time savers for me is after I set my pinch dimensions (on my head rotator) to give me more room to work the port I move on to the common wall and port to the tube. This all gives me better visual and max room. I then flip the head roof side down to raise the roof before I touch the floor. Through experience and testing I have a snap gauge that I use to set the roof dimensions where I want it. Keep this dimension handy as I use it on every valve size from 2.02-2.08. At this point I can work the floor some but I do my finish floor work after I have the head off my rotator. Take a set of 4-6 inch calipers and with the head on its end measure up through the valve seat and measure from the head deck to the top of the shortside. Keep notes of these measurements as you do your flow tests. I have a different height for each valve size and again have this number written down. I wish I could give you that dimension for the different valves sizes but I have 100’s of hours of porting and testing to get that number. My set I’m working for that hopefully will soon get on my duster are flowing 334cfm so far and have a totally different dimension. I didn’t reread this as I have some running to do so I hope it makes sense.
You have been more than generous. This will keep me busy (and awake at night) for months.
 
With certain heads, I often say “it’s all about the SSR”.
Really……..but also not really.

Imagine messing with a SB J head, T/A version.
Someone does half assed bowl job that leaves the SSR shaped like an abrupt cliff and it flows like [email protected]” lift, and the flow starts to drop of after that.
These heads have lots of material to play with at the PRP.
You can easily widen the port in that area by .125”+.
It won’t improve the flow at all, because the poorly shaped SSR is still just as bad as it was.
However, If you left the PRP alone and put a decent radius on the SSR and laid it back a little, there would be a pretty easy 10cfm gain.

Conversely, I have also see it where the SSR is working well, and the flow numbers are good, and the drop off at high flow is minimal.
Then you go in and open the pinch up “more”, and all of a sudden the peak flow is less, and the upper end of the curve falls of worse.
My take on that is the SSR can no longer manage the volume of air trying to get around the turn, even if it’s only in specific areas of the turn.

I’ll reference PBR’s tests on the TF190’s where he’s found inconsistencies from port to port, and he’s mentioned remedying some of that away with some minor tweaking.
Is that tweaking happening at the pinch, or the SSR?
I’m going to guess it’s the SSR, but I’m sure he’ll chime in.

I guess what I’m saying is, if the SSR doesn’t have the proper form to allow the flow column to stay attached, you can’t fix it by reworking other areas of the port.

The trick is…….knowing when the separation is caused by the shape of the floor, or by deficiencies in other areas of the port…….or both.
 
Last edited:
With certain heads, I often say “it’s all about the SSR”.
Really……..but also not really.

Imagine messing with a SB J head, T/A version.
Someone does half assed bowl job that leaves the SSR shaped like an abrupt cliff and it flows like [email protected]” lift, and the flow starts to drop of after that.
These heads have lots of material to play with at the PRP.
You can easily widen the port in that area by .125”+.
It won’t improve the flow at all, because the poorly shaped SSR is still just as bad as it was.
However, If you left the PRP alone and put a decent radius on the SSR and laid it back a little, there would be a pretty easy 10cfm gain.

Conversely, I have also see it where the SSR is working well, and the flow numbers are good, and the drop off at high flow is minimal.
Then you go in and open the pinch up “more”, and all of a sudden the peak flow is less, and the upper end of the curve falls of worse.
My take on that is the SSR can no longer manage the volume of air trying to get around the turn, even if it’s only in specific areas of the turn.

I’ll reference PBR’s tests on the TF190’s where he’s found inconsistencies from port to port, and he’s mentioned remedying so of it with some minor tweaking.
Is that tweaking happening at the pinch, or the SSR?
I’m going to guess it’s the SSR, but I’m sure he’ll chime in.

I guess what I’m saying is, if the SSR doesn’t have the proper form to allow the flow column to stay attached, you can’t fix it by reworking other areas of the port.

The trick is…….knowing when the separation is caused by the shape of the floor, or by deficiencies in other areas of the port…….or both.


The airspeed was 401fps pegged on my gauge so I mostly widened then reshaping. It still needs area so I am lowering the shortturn when I get time to touch them. It is way way higher than a stock Edelbrock or Speedmaster head.
 
Conversely, I have also see it where the SSR is working well, and the flow numbers are good, and the drop off at high flow is minimal.
Then you go in and open the pinch up “more”, and all of a sudden the peak flow is less, and the upper end of the curve falls of worse.
My take on that is the SSR can no longer manage the volume of air trying to get around the turn, even if it’s only in specific areas of the turn.
This is what I believe as well. Erland Cox has a good thread on Speed-talk about chamber stall. His theory is that the wake from the flow separation on the short side continues to get larger and shuts down the flow area. The 'wake' is actually the boundary layer growing larger and getting forced back up the port by the high pressure in the combustion chamber. I have verified the flow reversion and swirling with my testing with the pitot in the chamber.
 
PBR, I suspected you were going after the SSR in some way, and you’re finding gains without touching the PRP.
 
Last edited:
My concern with laying back the SS too much is the loss of torque. I'm sure you've seen Darin Morgan talk about a tall SS being good for torque and a low SS being good for HP. Would the SS be designed different for a street car than a race car? In general the answer would be yes, but the SBM heads don't play by the same rules as some of Morgan's stuff.
 
From what I understand, flow gains doesn’t trump port energy and efficiency. I believe that comes from trying to equalize port speeds throughout all areas of the port and trying to keep speeds around 300 for stockish size ports (with in reason for porting with available meat to port).
 
The heads I’m most familiar with, wrt SSR issues are BB 906 and 915 heads.

While I have seen very good flowing examples of those where the SSR hasn’t been lowered/laid back as much as I typically do(when heavily ported), I’ve never been able to duplicate those results.

On the other hand, I’ve had some ported examples show up here that had less that stellar flow, where I was able to pick up 20-30cfm by doing nothing but rework the SSR.
It’s pretty easy to end up with a big valve(where the whole intake runner has been ported) 906/915 intake port that’s stuck in the 240’s at .450-.500 lift, and either doesn’t show any gain beyond that, or the numbers go back down.
 
Last edited:
My concern with laying back the SS too much is the loss of torque. I'm sure you've seen Darin Morgan talk about a tall SS being good for torque and a low SS being good for HP. Would the SS be designed different for a street car than a race car? In general the answer would be yes, but the SBM heads don't play by the same rules as some of Morgan's stuff.

Remember my wording when it comes to short side. I lower then shape the short side. A lay back on a piss poor short side like the Edelbrock is a small radius. Now if we were playing with a W2 head or a Chevy Dart Platinum head we are talking another story. Both have beautiful short sides.
 
From what I understand, flow gains doesn’t trump port energy and efficiency. I believe that comes from trying to equalize port speeds throughout all areas of the port and trying to keep speeds around 300 for stockish size ports (with in reason for porting with available meat to port).
You are correct, maximum flow is not what is most important. It's more about velocity (from which port energy is derived) and shape. It would be nice to keep the port speeds around 300 fps average, but for a SM/Edelbrock style head on a large cubic inch motor, there is not enough area available to get the speeds that low.

The whole 'see how much flow I can get' thing is just a way to keep the testing interesting. You have to have a goal - something you are trying to achieve. That way you can look back at the end of the day and (sometimes) feel like you accomplished something.

Efficiency is a tough one to quantify. For instance - try to measure valve curtain area and really get it right.....
 
You are correct, maximum flow is not what is most important. It's more about velocity (from which port energy is derived) and shape. It would be nice to keep the port speeds around 300 fps average, but for a SM/Edelbrock style head on a large cubic inch motor, there is not enough area available to get the speeds that low.

The whole 'see how much flow I can get' thing is just a way to keep the testing interesting. You have to have a goal - something you are trying to achieve. That way you can look back at the end of the day and (sometimes) feel like you accomplished something.

Efficiency is a tough one to quantify. For instance - try to measure valve curtain area and really get it right.....

Wow. And Remember that I have boatloads of questions you guys already have answers to.
 
Wow. And Remember that I have boatloads of questions you guys already have answers to.
Don't be afraid to ask. It's how you learn. I'm by no means a guru like some of these other guys, but I'll help where I can.
 
Don't be afraid to ask. It's how you learn. I'm by no means a guru like some of these other guys, but I'll help where I can.
I’m not shy, I spout off plenty(god forgive me). I’m just reading and digesting.
 
I’m not shy, I spout off plenty(god forgive me). I’m just reading and digesting.

Look at lots of pictures and imagine it in 3D. I don’t make port molds as often as I should but probably will when I start my first LS head. I’m not sure you could even get the mold out of on of those damn heads. Lol
 
You are correct, maximum flow is not what is most important. It's more about velocity (from which port energy is derived) and shape. It would be nice to keep the port speeds around 300 fps average, but for a SM/Edelbrock style head on a large cubic inch motor, there is not enough area available to get the speeds that low.
From what I've been noticing from what I've been watching( I'm not a head porter: ), smallish engines at reasonable rpms eg.. 318-371 @ 5000-6500 rpms we have generally enough CSA available to us so it's of lesser concern, so getting the most flow for given size is generally beneficial, but once you start cranking up rpm especially with larger displacements gaining enough CSA seems to be the quest above all, like Darin's super stock hemi head that they spent countless hours in crazy mods not to gain even 1 cfm but to gain csa so they could shift like 500-600 rpms higher and probably spent $30,000 plus grand in the process.

To me it seems like CSA of the entire port and it's shape is most important and to slightly less extent cfm and volume are along for the ride plus we're generally not making full use of available cfm, 2-2.5 hp per cfm.
The whole 'see how much flow I can get' thing is just a way to keep the testing interesting. You have to have a goal - something you are trying to achieve. That way you can look back at the end of the day and (sometimes) feel like you accomplished something.

Efficiency is a tough one to quantify. For instance - try to measure valve curtain area and really get it right.....
My guess is the csa around the ssr will be the ultimate cork since it can only be modified so much, and design the rest the port to be in balance with that, unless there's a specific need to push everything to it's limit, I understand this just research phase looking for those limits. Keep up the good work.

I appreciate you sharing your porting journey with us, one day I hope to learn enough to take a crack at porting my heads.
 
One of the better “overachievers”(that I have first hand knowledge of) in the hp/cfm dept is actually my friends 350 Pontiac NHRA stock eliminator engine.
And it accomplished that while hampered with the factory intake/carb, and a .421” lift cam.

Sure, the power isn’t earth shattering, but neither is the head flow.
 
One of the better “overachievers”(that I have first hand knowledge of) in the hp/cfm dept is actually my friends 350 Pontiac NHRA stock eliminator engine.
And it accomplished that while hampered with the factory intake/carb, and a .421” lift cam.

Sure, the power isn’t earth shattering, but neither is the head flow.
It's crazy the things those guys can squeeze out of those cars.
 
-
Back
Top