Thanks, I have a couple of questions.
I am curious about the radius comment and want to understand it. Intuitively it seems like anything you can do to open up flow, bench or not, would help. What causes the problem wrt a radius?
Top cut, check. This makes sense.
Can you talk more about the relationship between the valve and port angle regarding seat angles etc?
Throat % I think I understand a the size of the opening below the valve compared to the contact ring of the valve+seat? I see 85-90% tossed around depending on the goal.
Thanks again.
TL;TR…a radius flows better in reverse than a multi-angle valve job and that kills power. For the long explanation read the rest.
When looking at numbers coming off a flow bench, you have to consider its limits. Like any tool, a flow bench has limits.
Also, just looking at flow numbers as the sole source of information can get you down a power dropping rabbit hole.
Flow numbers sell heads, but horsepower wins races.
Having said all of that, I suppose we need to discuss what some of the limits of a flow bench are.
1. A flow bench measures air flow in one direction at a time.
2. A flow bench flows at a fixed test pressure (a depression in inches of water).
3. For the most part, you test on a flow bench with only one valve open.
So what’s wrong with flowing in one direction? Because in a running engine the flow is never in one direction. If you haven’t seen the Jon Kasse video where he sticks his finger in the tunnel ram engine on they dyno, you should do a search and watch it.
What’s wrong with flowing at a constant test pressure? Same as above. A running engine never sees a constant pressure drop across the valve. Some ports can see upwards of 150 inches of water when the valve breaks the seat!! And the port may only see 8-10 inches of water at maximum lift. So measuring at 28 inches of water is a compromise.
What’s the big deal about testing with one valve open? Because at overlap, when both valves are open you can make some power BUT you can piss away a bunch of power. It’s difficult to measure flow with both valves open. I tried many times to do it. All without success.
What does all this mean? When you get a cylinder head flowed and you get a flow sheet with the flow numbers on it, you have no idea how well the port flows at any other pressure drop than 28 inches, or whatever test pressure the operator uses. Or used.
A port that looks good at 10 inches of water may suck buttermilk at 28 inches and at 40 inches it may kick ***. But you dot know that information because either the port wasn’t tested at those different depressions or maybe they were but you weren’t given that data.
I didn’t settle on 50 degree valve jobs by accident. Ok, that’s not exactly true. I was lucky enough to work on some really cool stuff in my career and my exposer to steeper than 45 degree seats came about because of that work. In mid 1999 I found out a cylinder head I was working on had 55 degree seats. From that point forward I have tested every conceivable seat angle from 30 to 55 and for most street/strip stuff a 50 is about the best.
If you test a 50 degree valve job at 28 inches you’ll quickly find that that is LOSES a significant amount of flow to a 45. BUT, if you run the test pressure up to say…48 inches (as high as my bench would go at that time) you’d see that the 50 was kicking ***.
And 48 inches is far LESS than what the port “sees” at low lift.
There is math (Super Flow made up a chart) to calculate what the flow should be if you want to test a port at say, 28 inches and then at 48 inches. In the real world, you never get the exact mathematical gain going from 28 to 48 inches of water because of internal friction in the port and other things I can’t think of now.
So, if you flow a head at 28 and then at 48 what you want to see is the least amount loss at 48 inches compared to the math.
As an example (because I can remember the math lol) let’s say you are testing at 25 inches and you have the port about done. Now you run the test pressure to 28 inches and you see bigger numbers but how does the math line up?
I know (from memory) that the difference between 25 and 28 inches is 6%. IOW’s, in a perfect world (the world is far from perfect) the port should flow 6% more at 28 inches than it did at 25 inches.
For math’s sake, let’s say you ported a head and at 25 inches it flows 300 CFM (at peak). That means mathematically the port should flow 318 CFM at 28 inches. And it won’t, because of friction and such. What you want to see is the least amount of difference in the what the math says it should flow and what the port really flows.
And now we go down another rabbit hole. Because we are trying to reduce the losses from 25 to 28 inches we do wacky stuff like polish the port to a high sheen. And the port flows better. BUT, on the dyno and in the car the engine is down on power. What happened?
Well, we let flow over rule all else. We gained a bit of flow at the cost of horsepower because while the highly polished port flowed more, that finish made the engine use a bunch of fuel (look at BSFC numbers and you’ll see it) and the engine didn’t convert that fuel into useful work. The rough finish kept more of the fuel atomized in the port and when it hit the cylinder the fuel/air mixture was better prepared to be vaporized and that makes power.
Thats one example of using flow bench numbers as your only basis of determining if a port is good or not.
Sooooooooo, now we get to the radius valve job for the intake port. And why it’s a guaranteed powa killa.
If you consider (and test) air flow in only one direction you never get close to duplicating what happens in the port. There is always air/fuel moving in both directions in the port.
I always flowed the port in BOTH directions (I learned this the hard way after I bought my bench and I cost me a couple of nights of testing until I figured out what I was doing wrong).
Kind of like the example above with surface finish, flowing the port in both directions can teach you a lot.
One of the things you learn is a port that flows as good in reverse as it flows in its “normal” direction (in this case we are talking about the intake port so we pull the air from from the manifold face so reverse flow would be pushing air from the chamber out of the port to the manifold) that port will lose power and burn more fuel for that less power every single time.
The question is why? It’s a couple of things that I can think of. One reason is (talking about carb’d stuff ATM) that at the booster where fuel is added to the air you have air moving in two directions. The “normal“ direction is air moving from the filter (or scoop) through the carb to the cylinder and “reverse” is air moving from the chamber back up the runner to the booster and the air filter (or scoop).
What does all that mean? The booster is dumb. It doesn’t know which way the air is moving through it. It just knows there is a pressure drop across it so it adds fuel regardless of which direction it’s going. Do you see the issue yet?
Air moves through the booster in the normal direction and it picks up fuel from the booster. Now you have some reverse flow and that same air moves back through the booster in the reverse direction. The booster, being dumb doesn’t know the air is going the wrong way so it adds fuel. The reverse part of the event stops and the air/fuel moves back to the normal direction and as that same air/fuel moves back through the booster and it picks up fuel for a THIRD time.
You have now added DOUBLE the fuel because the booster doesn’t know which way the air is moving. And you have all that extra fuel doing nothing to make power. That shows up in higher BSFC numbers.
So reducing reverse flow is an easy way to make power and increase the efficiency of the engine. So how does all this make a difference between a multi angle valve job compared to the radius valve job on the intake?
Reverse flow though the booster.
A radius seat on the intake port will ENCOURAGE reverse flow. And that’s bad.
Let’s say we have our 300 CFM port with a 4 angle valve job and we test it in reverse (making up numbers here) and in reverse it flows 250 CFM. Now we cut a radius valve job on it and we get 315 CFM. We are doing hand stands and cart wheels across the shop, and I grab the phone and call Hot Rod magazine and tell them I’m about to rotate the earth and I want some magazine time. No one else in the country is getting that much CFM out of these ports. So we call it a day, head to the bar and tie on a big one.
The next day we head to the shop, hung over like a freight train and strap this world beater to the dyno. After 20 pulls we see it’s a fuel burning dong beater and would be lucky to pull the hat off your head. So what went wrong? We didn’t test the port in reverse.
Had we done that, we would see that we gained 15 CFM in the normal direction BUT we also now flow 300 CFM in REVERSE. We picked up 50 (FIFTY) CFM in reverse flow to gain 15 CFM in normal flow. And that’s a power loser.
I‘d rather lose 10 CFM in normal flow to kill that same CFM (or more) in reverse because I know that will make more power.
Once I learned this (and how to “read“ intake and exhaust ports after running the engine) I always looked at reverse flow.
It’s damned hard to take a guys cylinder heads that flow X CFM, fix them and now they flow Y CFM (where Y is LESS than X) and telling him your ports will make more power with LESS flow. It’s near impossible to sell ports like that when the entire market is brainwashed to look at CFM alone.
The other issue is at overlap, when both valves are off the seat. You want the header to tug on the intake port (and by doing that pulling harder on the booster so you can run LESS jet for the same or more power) to get more of the fresh charge of air/fuel into the chamber.
But, we have the high flowing (in both directions) ports now sending that air/fuel mixture backwards up the port rather than getting it into the chamber and you get the power loss and efficiency loss from that reverse flow.
I know that’s a long explanation but it’s the only way I know to explain it so it makes sense.
I constantly harp on not buying heads based on CFM (carbs too) because it’s just not that simple.
I almost forgot that the sharp angles of a multi-angle valve job also work to re-atomize the fuel as it hits the valve job. You lose that with a radius.
I hope this makes sense.
RB
Seat grinders have been used for a century and the right stones and the right angles in the right hands can produce good results. The Serdi style carbide cutters really do a nice job creating a smooth radius instead of multiple angles like a stone.
