Cylinder Heads - What Matters Most?
05-31-2019, 04:57 PM
#265
Some rambling thoughts..
The purpose of a cylinder head is to trap air inside the combustion chamber. They are not steady state devices, which is what a flow bench is.
The act of trapping air in the combustion chamber requires looking at the entire induction and exhaust together as a system, along with cam timing and RPM.
The flow of air in an engine is done by pressure differential. The piston moving downward creates a low pressure in the cylinder, and the job of the induction system is to provide a high pressure wave at the intake valve at key moments during the induction cycle - first on the piston down-stroke, and then also during intake valve close when the piston is coming back up.
The size and length of the induction tract changes the timing of the pressure wave signal, as well as the inertia of the air in the induction tract. It also affects the strength of those pressure waves, longer = stronger...momentum. The speed of the pressure wave is always the speed of sound at the air temp in the intake tract.
As the air starts flowing, a low pressure wave travels up the runner and hits the plenum and then the plenum starts back filling the runner. The air in the runner is essentially what I think about as "readily available" as it has relatively low intertia compared to the plenum. Once that runner is depleted you need to wait for air to arrive from the plenum. Short, large runners don't work effectively at low RPM because the engine isn't creating enough demand to move the huge mass of air in the plenum. However, once the depression in the cylinder increases those short fat runners make way for a lot of airflow. The time it takes for the volume of air to reach the valve is greatly reduced, which is what you need in high rpm situations. Air does no good if it just slams into a closed intake valve.
If the port is "lazy" it means that the air has high inertia, and there's not enough time for significant air to actually flow into the combustion chamber due to the differential pressure. Also, it could refer to the piston pushing the intake charge back into the intake tract effectively losing fuel and air that could be offering useful work. High flow at low RPM can work against you with big cams...it flows both ways.
Steady state flow numbers on a bench are good for R&D purposes, but pretty worthless for comparing heads across benches or really actually understanding what is going on inside the engine when it is running. I'm unsure if there are any flow benches in the world that can actually generate the correct depression to mimic the draw of a piston, and I don't know of any that can create the transient conditions actually seen with the valve opening and closing along with pressure rising and falling behind the valve.
I'd say the most critical parts of a cylinder head are going to be the short side turn, and then very similar in importance the actual valve job and entry into the cylinder. The highest speeds are going to be seen at the entry to the cylinder, and then immediately the air expands dramatically and loses all of its velocity. It basically shoots out of a nozzle and stalls in a giant volume of air.
The short side is the most difficult area for the intake charge to traverse as you have a pretty massive amount of momentum piling straight ahead down the runner and you need to somehow convince it to turn.
The min CSA is usually at the restriction in the port at the valve stem. The throat of the valve seat is very similar in size to the min CSA, and is usually 85% +/- the diameter of the valve. The CSA of the port pretty much gradually expands outward to the intake flange and will set the volume of the port and is probably the least important part of any of it - yet is pretty much what sets the CC's of the runner just by sheer percentage of the intake tract it makes up.
I feel like after the valve size and throat diameter, the short side is developed, and the intake flange is known the rest of the port sort of falls into place.
The purpose of a cylinder head is to trap air inside the combustion chamber. They are not steady state devices, which is what a flow bench is.
The act of trapping air in the combustion chamber requires looking at the entire induction and exhaust together as a system, along with cam timing and RPM.
The flow of air in an engine is done by pressure differential. The piston moving downward creates a low pressure in the cylinder, and the job of the induction system is to provide a high pressure wave at the intake valve at key moments during the induction cycle - first on the piston down-stroke, and then also during intake valve close when the piston is coming back up.
The size and length of the induction tract changes the timing of the pressure wave signal, as well as the inertia of the air in the induction tract. It also affects the strength of those pressure waves, longer = stronger...momentum. The speed of the pressure wave is always the speed of sound at the air temp in the intake tract.
As the air starts flowing, a low pressure wave travels up the runner and hits the plenum and then the plenum starts back filling the runner. The air in the runner is essentially what I think about as "readily available" as it has relatively low intertia compared to the plenum. Once that runner is depleted you need to wait for air to arrive from the plenum. Short, large runners don't work effectively at low RPM because the engine isn't creating enough demand to move the huge mass of air in the plenum. However, once the depression in the cylinder increases those short fat runners make way for a lot of airflow. The time it takes for the volume of air to reach the valve is greatly reduced, which is what you need in high rpm situations. Air does no good if it just slams into a closed intake valve.
If the port is "lazy" it means that the air has high inertia, and there's not enough time for significant air to actually flow into the combustion chamber due to the differential pressure. Also, it could refer to the piston pushing the intake charge back into the intake tract effectively losing fuel and air that could be offering useful work. High flow at low RPM can work against you with big cams...it flows both ways.
Steady state flow numbers on a bench are good for R&D purposes, but pretty worthless for comparing heads across benches or really actually understanding what is going on inside the engine when it is running. I'm unsure if there are any flow benches in the world that can actually generate the correct depression to mimic the draw of a piston, and I don't know of any that can create the transient conditions actually seen with the valve opening and closing along with pressure rising and falling behind the valve.
I'd say the most critical parts of a cylinder head are going to be the short side turn, and then very similar in importance the actual valve job and entry into the cylinder. The highest speeds are going to be seen at the entry to the cylinder, and then immediately the air expands dramatically and loses all of its velocity. It basically shoots out of a nozzle and stalls in a giant volume of air.
The short side is the most difficult area for the intake charge to traverse as you have a pretty massive amount of momentum piling straight ahead down the runner and you need to somehow convince it to turn.
The min CSA is usually at the restriction in the port at the valve stem. The throat of the valve seat is very similar in size to the min CSA, and is usually 85% +/- the diameter of the valve. The CSA of the port pretty much gradually expands outward to the intake flange and will set the volume of the port and is probably the least important part of any of it - yet is pretty much what sets the CC's of the runner just by sheer percentage of the intake tract it makes up.
I feel like after the valve size and throat diameter, the short side is developed, and the intake flange is known the rest of the port sort of falls into place.
05-31-2019, 05:14 PM
#266
Moderator
Thread Starter
iTrader: (4)
Join Date: Jul 2014
Location: My own internal universe
Posts: 10,452
Received 1,852 Likes
on
1,152 Posts
Well said spanks!
Where those pressure waves come in is peak VE or peak tq. From the mouth of the runner to the combustion chamber there is an ideal open and close frequency that creates peak cylinder fill. Change the runner length you change that frequency. Change the csa and you change that frequency. Below that frequency (lower rpm) you get reversion. Above it (higher rpm) you lose time to get air into the cylinder.
Where those pressure waves come in is peak VE or peak tq. From the mouth of the runner to the combustion chamber there is an ideal open and close frequency that creates peak cylinder fill. Change the runner length you change that frequency. Change the csa and you change that frequency. Below that frequency (lower rpm) you get reversion. Above it (higher rpm) you lose time to get air into the cylinder.
05-31-2019, 09:28 PM
#267
TECH Veteran
Darth~ I been thinking about some things all day and im generally speaking out loud here. Have you ever paid any attention to some of the rwhp dyno numbers that some of the LS3 headed strokers make? Its a few on board here with track results and dyno sheets. MOST of them fall short of the 500rwhp mark. Not go call the member name buts its this one on board here running a 40Xci, T56, CNC ported LS3 heads, LS3 intake , cam motion rectangle port stroker cam..... anyway the car makes less than 480rwhp on a mustang dyno and the car cant seem to get past 123 mph trap speed. Wonder whats the problem?
05-31-2019, 10:50 PM
#269
I can offer one LS3 data point. I've posted my dyno results here before for a couple of combos now. My LS3-based setup was a 434ci, LLSR from Cam Motion, and TFS GenX castings which had a semi-custom port from Frankenstein Racing Heads (now FED). Chris Frank and I had a few discussions about what I wanted, and we ended up with a 273cc intake port and a 2.165" intake valve. I was using a ported Victor Jr. single plane intake (which is different than many/most LS combinations which are discussed here). If you want the rest of the combination, feel free to search my username.
Anyway, that combination made 580 rwhp through a T56 Magnum and Ford 9". Absolutely zero dyno optimizations, it's a full-on Pro-Touring car. Torque was right around 500 - 505 RW as I recall. I never took that car to a drag strip so I don't have track data to share, but the car really did run well for what I do.
I've currently rebuilt the setup with the MMS 265 heads and still a ported single-plane intake. I haven't run it on the same chassis dyno yet (plan to soon), but I did engine dyno this one. I've had in my head that I wanted an LS7-based combination for a long time. As many talk about here, I felt with my larger bore I could take advantage of a larger valve and the seemingly better intake port track of the LS7. I'm very curious to see how the results compare on the same dyno. Will be telling for me whether the large cash investment was worth it...
I was always pretty impressed with the FRH LS3 head combination that I had. I'll be kicking myself if I don't see some gains.
Anyway, that combination made 580 rwhp through a T56 Magnum and Ford 9". Absolutely zero dyno optimizations, it's a full-on Pro-Touring car. Torque was right around 500 - 505 RW as I recall. I never took that car to a drag strip so I don't have track data to share, but the car really did run well for what I do.
I've currently rebuilt the setup with the MMS 265 heads and still a ported single-plane intake. I haven't run it on the same chassis dyno yet (plan to soon), but I did engine dyno this one. I've had in my head that I wanted an LS7-based combination for a long time. As many talk about here, I felt with my larger bore I could take advantage of a larger valve and the seemingly better intake port track of the LS7. I'm very curious to see how the results compare on the same dyno. Will be telling for me whether the large cash investment was worth it...
I was always pretty impressed with the FRH LS3 head combination that I had. I'll be kicking myself if I don't see some gains.
06-01-2019, 02:50 AM
#270
Darth~ I been thinking about some things all day and im generally speaking out loud here. Have you ever paid any attention to some of the rwhp dyno numbers that some of the LS3 headed strokers make? Its a few on board here with track results and dyno sheets. MOST of them fall short of the 500rwhp mark. Not go call the member name buts its this one on board here running a 40Xci, T56, CNC ported LS3 heads, LS3 intake , cam motion rectangle port stroker cam..... anyway the car makes less than 480rwhp on a mustang dyno and the car cant seem to get past 123 mph trap speed. Wonder whats the problem?
It could be one of many things.
My LS 408 stroker made 505rwhp with a mild 236/244 114 LSA camshaft, through 3:91 gears and a 3200rpm stall.
I'm fitting a larger camshaft and can't wait to see what it does with this bad boy.
06-01-2019, 05:58 AM
#271
TECH Veteran
Depends how they are set up and who does the work etc..
It could be one of many things.
My LS 408 stroker made 505rwhp with a mild 236/244 114 LSA camshaft, through 3:91 gears and a 3200rpm stall.
I'm fitting a larger camshaft and can't wait to see what it does with this bad boy.
It could be one of many things.
My LS 408 stroker made 505rwhp with a mild 236/244 114 LSA camshaft, through 3:91 gears and a 3200rpm stall.
I'm fitting a larger camshaft and can't wait to see what it does with this bad boy.
06-01-2019, 07:35 AM
#272
06-01-2019, 08:25 AM
#273
Moderator
Thread Starter
iTrader: (4)
Join Date: Jul 2014
Location: My own internal universe
Posts: 10,452
Received 1,852 Likes
on
1,152 Posts
Originally Posted by Tuskyz28
Darth~ I been thinking about some things all day and im generally speaking out loud here. Have you ever paid any attention to some of the rwhp dyno numbers that some of the LS3 headed strokers make? Its a few on board here with track results and dyno sheets. MOST of them fall short of the 500rwhp mark. Not go call the member name buts its this one on board here running a 40Xci, T56, CNC ported LS3 heads, LS3 intake , cam motion rectangle port stroker cam..... anyway the car makes less than 480rwhp on a mustang dyno and the car cant seem to get past 123 mph trap speed. Wonder whats the problem?
Stock intake wants to force peak rpm to 6300. You can cheat that a little with cam but not much. Longer stroke pulls the peak down. Likely closer to 6100. First thing on my list would be a shorter runner intake like fast mid length. Owner likely will not do it to avoid losing low end torque.
06-01-2019, 11:23 AM
#274
That is exactly what I mean about underwhelming strokers. Not knowing much else, here is my guess:
Stock intake wants to force peak rpm to 6300. You can cheat that a little with cam but not much. Longer stroke pulls the peak down. Likely closer to 6100. First thing on my list would be a shorter runner intake like fast mid length. Owner likely will not do it to avoid losing low end torque.
Stock intake wants to force peak rpm to 6300. You can cheat that a little with cam but not much. Longer stroke pulls the peak down. Likely closer to 6100. First thing on my list would be a shorter runner intake like fast mid length. Owner likely will not do it to avoid losing low end torque.
My engine peaked at 6100rpm (with current cam), held peak power to 6300rpm and then started falling off.
This is with the FAST 102mm manifold.
We spoke about mid length runners but I am not willing to sacrifice the torque in the low and mid range although I will get the 7000rpm+ range I am after.
I had another member from here who messaged me and we had a good discussion about this.
This gentleman has an LS3 427 with the FAST 102mm intake and CNC LS3 heads.
When he changed to the mid length runners, he said it made his car feel at least 500lb heavier with quite a substantial loss of torque off idle-6000rpm.
From 6000rpm+ the torque evened out and made a little more but of course with a few extra ponies at 7200rpm.
This was on a larger engine than mine too and the torque loss was significant. On mine, the losses would be higher.
This is another reason why it's holding me back and it's best to stick with the long runners on the street.
06-01-2019, 11:48 AM
#275
Darth, this sounds very similar to my engine.
My engine peaked at 6100rpm (with current cam), held peak power to 6300rpm and then started falling off.
This is with the FAST 102mm manifold.
We spoke about mid length runners but I am not willing to sacrifice the torque in the low and mid range although I will get the 7000rpm+ range I am after.
I had another member from here who messaged me and we had a good discussion about this.
This gentleman has an LS3 427 with the FAST 102mm intake and CNC LS3 heads.
When he changed to the mid length runners, he said it made his car feel at least 500lb heavier with quite a substantial loss of torque off idle-6000rpm.
From 6000rpm+ the torque evened out and made a little more but of course with a few extra ponies at 7200rpm.
This was on a larger engine than mine too and the torque loss was significant. On mine, the losses would be higher.
This is another reason why it's holding me back and it's best to stick with the long runners on the street.
My engine peaked at 6100rpm (with current cam), held peak power to 6300rpm and then started falling off.
This is with the FAST 102mm manifold.
We spoke about mid length runners but I am not willing to sacrifice the torque in the low and mid range although I will get the 7000rpm+ range I am after.
I had another member from here who messaged me and we had a good discussion about this.
This gentleman has an LS3 427 with the FAST 102mm intake and CNC LS3 heads.
When he changed to the mid length runners, he said it made his car feel at least 500lb heavier with quite a substantial loss of torque off idle-6000rpm.
From 6000rpm+ the torque evened out and made a little more but of course with a few extra ponies at 7200rpm.
This was on a larger engine than mine too and the torque loss was significant. On mine, the losses would be higher.
This is another reason why it's holding me back and it's best to stick with the long runners on the street.
06-01-2019, 12:49 PM
#277
TECH Regular
It's all in what you're after and the combination you need to use to get you there. In 2017 my 97 S10 ran 11.18 with a mild LQ4, G5x3 cam, 3500 stall, Turbo 350, 4.10 gears, TBSS intake. So in 2018 I built a small bore 421 stroker with stock 823 heads, Texas speed cam, and rod/mod ported LS3 intake and it only went 10.80's, I was disappointed to say the least but I knew from the dyno what the problem was, the engine was out of air at only 5800 rpm, so I took some advice from NavyBlue and switched to a Fast 102 mid runner intake, ported 821 heads, Cam motion cam, FTI 4500 stall, 4.30 gears, this combination peaked at 6800 and ended up going a 10.33 shifting at 7000 and coming through the traps at 7000. This year I replaced the 821 heads with AFR Mongoose heads and custom cam from Tony Mamo, it picked up nicely on the dyno, but only went 10.26 which isn't much faster than before, but to keep things in perspective I've put in a roll cage which added about 100 pounds, plus I switched from the turbo 350 to a FTI powerglide. My truck is heavy with full interior,heater,wipers,and stereo still intact and would have run better times with the 3 speed but it was terribly inconsistent off the line because my starting line ratio was way too high, and the powerglide allows me the consistency I need for racing. There is no 1 recipe that fits every situation, you have to decide your goals are and experiment with different combinations to meet them. The next experiment for me will be to install a set of short runners in my fast intake, my thoughts are that they might pick me up due to my high piston speed and those hungry Mongoose heads. If they don't work for me, they are easy enough to swap back out. I've also ordered some double adjustable shocks....it's time to ditch the bargain basement 3-ways...lol.
Last edited by BigDaddy97; 06-01-2019 at 01:21 PM.
06-01-2019, 12:51 PM
#278
Moderator
Thread Starter
iTrader: (4)
Join Date: Jul 2014
Location: My own internal universe
Posts: 10,452
Received 1,852 Likes
on
1,152 Posts
I run 4.30 gears and a sniper intake. I have no clue what rpm I peak torque but peak power at 7000.
So much torque it is hard to launch
Edit - you might find it is easier to launch on short runners due to some reduction in midrange torque.
So much torque it is hard to launch
Edit - you might find it is easier to launch on short runners due to some reduction in midrange torque.
06-01-2019, 12:59 PM
#279
TECH Regular
I've looked at the results from that LS3 intake shootout and I'm liking that Sniper Low Pro intake. It makes great power and would clear everything in my engine bay.
06-01-2019, 01:12 PM
#280
Good info.
How can a set of LS3's run out of air at 5800rpm on a 421?
There have been many tests that even with stock heads, they will pull to 6800-7000rpm without an issue if you have enough camshaft.
Brian Tooley's stroker camshaft pulls to at least 6500rpm in a 416 LS3 with stock heads and intake.
This is a 239/254 114 LSA +3
What are the specs of the texas speed camshaft you had in there?
If you have an IVC of 50 and an EVO of 64 you will pull to 6500rpm quite easily.
The mid runner will naturally raise where the engine will peak.
Check out the dyno sheet of the Brian Tooley stroker camshaft on the engine combination I mentioned.