Theory of 4 valve poly quad cylinder heads.

General engine tech -- Drag Racing to Circle Track

Moderator: Team

4vpc
Pro
Pro
Posts: 337
Joined: Thu Jan 21, 2016 6:26 pm
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by 4vpc »

You need to buy and read his book then you won't have to guess or assume, it's not expensive. I think it's a bit cheap to post up pictures from it.
The car featured with Polyquad was a Mitsubishi Evo with VVT. I think the idea will have come from an existing design and I did rebuild a 16v Renault diesel once which had two completely different port lengths on the inlet. One port came traditionally through the side, but the other came through the top of the head. Also the three pulleys running the cam timing belt were of the same diameter, but these in turn were driven by gearwheels, it was a really interesting design, although as I haven't worked out how to post up a pic yet i'll leave that there.
It's a shame no-one else has ever publicly tested to see if PQ works or not, as far as I am aware anyone can do it as long as it's not sold.
There is no S on the end of RPM.
hoffman900
HotPass
HotPass
Posts: 3445
Joined: Sat Feb 23, 2013 5:42 pm
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by hoffman900 »

https://www.hondarandd.jp/

Sign up (it’s free) and search “tumble” to get an awesome education.
-Bob
4vpc
Pro
Pro
Posts: 337
Joined: Thu Jan 21, 2016 6:26 pm
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by 4vpc »

ptuomov wrote: Tue Sep 10, 2019 7:33 pm Stuff
On a similar note where can I find your turbo squish pad thread? I think you started one or something similar. I came across something of vague interest the other day, but don't know whether it was noted there or not.
There is no S on the end of RPM.
hoffman900
HotPass
HotPass
Posts: 3445
Joined: Sat Feb 23, 2013 5:42 pm
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by hoffman900 »

Honda’s steady state tumble rig.
You do not have the required permissions to view the files attached to this post.
-Bob
4vpc
Pro
Pro
Posts: 337
Joined: Thu Jan 21, 2016 6:26 pm
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by 4vpc »

LoganD wrote: Tue Sep 10, 2019 10:10 am You don't want swirl in a gasoline spark ignition engine. If you're seeing combustion efficiency improvements with the introduction of swirl then the system simply didn't have enough charge motion to begin with. Tumble is what you want in a gasoline SI engine. High swirl rates produce fuel/air separation and move the fuel away from the spark plug (centripetal acceleration) producing a localized lean zone by the spark plug....and that's the exact opposite of what you want.

This is why you've seen the move away from steep valve angles in 4-valve cylinder heads to much shallower valve angles without a major change in port angle. The amount of tumble in a new 4-valve cylinder head (anything post 2018) is crazy, and this is how they're getting 32 bar BMEP on 91 octane.

This WAS a debate for a long time, but it turned out that the steep valve angles and large ports in 90s 4-valve heads, particularly those from Honda and Mitsubishi, just didn't have any charge motion at low pressure drops. Modern heads to not have this issue. You do sacrifice some flow for any charge motion, so some manufacturers have compromised with moderate valve angles and some sort of flap that will help charge motion at low pressure drops. Ford did this with the Coyote V8, it has "CMCV" plates or charge motion control valves.

We measure this now with a flow index. We have a tumble meter on the flow bench and we rate the heads by (flow*tumble)/valve size. You're looking to maximize tumble while still utilizing as much of the valve curtain area as possible with a given valve size. Mercedes are the absolute leaders in this, it's borderline witchcraft how good they've gotten at balancing this.
If it wasn't needed the manufacturers wouldn't be building it in.
The latest engines have DI so go by different rules to the port injected ones we are discussing.
There is no S on the end of RPM.
hoffman900
HotPass
HotPass
Posts: 3445
Joined: Sat Feb 23, 2013 5:42 pm
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by hoffman900 »

4vpc wrote: Wed Sep 11, 2019 4:23 am
LoganD wrote: Tue Sep 10, 2019 10:10 am You don't want swirl in a gasoline spark ignition engine. If you're seeing combustion efficiency improvements with the introduction of swirl then the system simply didn't have enough charge motion to begin with. Tumble is what you want in a gasoline SI engine. High swirl rates produce fuel/air separation and move the fuel away from the spark plug (centripetal acceleration) producing a localized lean zone by the spark plug....and that's the exact opposite of what you want.

This is why you've seen the move away from steep valve angles in 4-valve cylinder heads to much shallower valve angles without a major change in port angle. The amount of tumble in a new 4-valve cylinder head (anything post 2018) is crazy, and this is how they're getting 32 bar BMEP on 91 octane.

This WAS a debate for a long time, but it turned out that the steep valve angles and large ports in 90s 4-valve heads, particularly those from Honda and Mitsubishi, just didn't have any charge motion at low pressure drops. Modern heads to not have this issue. You do sacrifice some flow for any charge motion, so some manufacturers have compromised with moderate valve angles and some sort of flap that will help charge motion at low pressure drops. Ford did this with the Coyote V8, it has "CMCV" plates or charge motion control valves.

We measure this now with a flow index. We have a tumble meter on the flow bench and we rate the heads by (flow*tumble)/valve size. You're looking to maximize tumble while still utilizing as much of the valve curtain area as possible with a given valve size. Mercedes are the absolute leaders in this, it's borderline witchcraft how good they've gotten at balancing this.
If it wasn't needed the manufacturers wouldn't be building it in.
The latest engines have DI so go by different rules to the port injected ones we are discussing.
Not true. The high performance Superbike engines do not have DI. Those applications are 12.5:1+ compression on pump gas and make north of 3hp/ci from the factory. no swirl to speak of, all tumble.

Read white papers on the Honda tech site. The manufacturers not only tune the axis of tumble, but also direct it to the center of the bore via the piston top.
-Bob
User avatar
ptuomov
Guru
Guru
Posts: 3587
Joined: Fri Aug 07, 2009 3:52 am
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by ptuomov »

hoffman900 wrote: Wed Sep 11, 2019 6:36 am
4vpc wrote: Wed Sep 11, 2019 4:23 am
LoganD wrote: Tue Sep 10, 2019 10:10 am You don't want swirl in a gasoline spark ignition engine. If you're seeing combustion efficiency improvements with the introduction of swirl then the system simply didn't have enough charge motion to begin with. Tumble is what you want in a gasoline SI engine. High swirl rates produce fuel/air separation and move the fuel away from the spark plug (centripetal acceleration) producing a localized lean zone by the spark plug....and that's the exact opposite of what you want.

This is why you've seen the move away from steep valve angles in 4-valve cylinder heads to much shallower valve angles without a major change in port angle. The amount of tumble in a new 4-valve cylinder head (anything post 2018) is crazy, and this is how they're getting 32 bar BMEP on 91 octane.

This WAS a debate for a long time, but it turned out that the steep valve angles and large ports in 90s 4-valve heads, particularly those from Honda and Mitsubishi, just didn't have any charge motion at low pressure drops. Modern heads to not have this issue. You do sacrifice some flow for any charge motion, so some manufacturers have compromised with moderate valve angles and some sort of flap that will help charge motion at low pressure drops. Ford did this with the Coyote V8, it has "CMCV" plates or charge motion control valves.

We measure this now with a flow index. We have a tumble meter on the flow bench and we rate the heads by (flow*tumble)/valve size. You're looking to maximize tumble while still utilizing as much of the valve curtain area as possible with a given valve size. Mercedes are the absolute leaders in this, it's borderline witchcraft how good they've gotten at balancing this.
If it wasn't needed the manufacturers wouldn't be building it in.
The latest engines have DI so go by different rules to the port injected ones we are discussing.
Not true. The high performance Superbike engines do not have DI. Those applications are 12.5:1+ compression on pump gas and make north of 3hp/ci from the factory. no swirl to speak of, all tumble.

Read white papers on the Honda tech site. The manufacturers not only tune the axis of tumble, but also direct it to the center of the bore via the piston top.
There’s that SAE paper that relates the intake port angle to the exhaust valve angle for efficient tumble. Basically, the closed exhaust valve face needs to act as an efficient diffuser and flow guide to the flow out of the intake valve. The intake valve angle itself is less important, it’s the port angle that matters the most.

This is one of the reasons why I believe that maximizing flow bench flow alone in a four valve head isn’t what makes the most power. One also needs to pay attention to the tumble wheel.

I’m not sure about this but I believe that with a turbo engine more tumble isn’t always automatically better. Even if faster average burn would always be better, there’s also the variance of burn speeds that one needs to worry about, especially when knock limited on pump gas.

The piston guiding effects you speak of are related to high compression motors?
Paradigms often shift without the clutch -- https://www.youtube.com/watch?v=cxn-LxwsrnU
https://www.instagram.com/ptuomov/
Put Search Keywords Here
User avatar
ptuomov
Guru
Guru
Posts: 3587
Joined: Fri Aug 07, 2009 3:52 am
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by ptuomov »

4vpc wrote: Tue Sep 10, 2019 8:01 pm
ptuomov wrote: Tue Sep 10, 2019 7:33 pm Stuff
On a similar note where can I find your turbo squish pad thread? I think you started one or something similar. I came across something of vague interest the other day, but don't know whether it was noted there or not.
I’ve bern rambling for a while:

viewtopic.php?f=1&t=49232
viewtopic.php?f=15&t=44789
viewtopic.php?f=1&t=36031
Paradigms often shift without the clutch -- https://www.youtube.com/watch?v=cxn-LxwsrnU
https://www.instagram.com/ptuomov/
Put Search Keywords Here
hoffman900
HotPass
HotPass
Posts: 3445
Joined: Sat Feb 23, 2013 5:42 pm
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by hoffman900 »

ptuomov wrote: Wed Sep 11, 2019 7:57 am
hoffman900 wrote: Wed Sep 11, 2019 6:36 am
4vpc wrote: Wed Sep 11, 2019 4:23 am

If it wasn't needed the manufacturers wouldn't be building it in.
The latest engines have DI so go by different rules to the port injected ones we are discussing.
Not true. The high performance Superbike engines do not have DI. Those applications are 12.5:1+ compression on pump gas and make north of 3hp/ci from the factory. no swirl to speak of, all tumble.

Read white papers on the Honda tech site. The manufacturers not only tune the axis of tumble, but also direct it to the center of the bore via the piston top.
There’s that SAE paper that relates the intake port angle to the exhaust valve angle for efficient tumble. Basically, the closed exhaust valve face needs to act as an efficient diffuser and flow guide to the flow out of the intake valve. The intake valve angle itself is less important, it’s the port angle that matters the most.

This is one of the reasons why I believe that maximizing flow bench flow alone in a four valve head isn’t what makes the most power. One also needs to pay attention to the tumble wheel.

I’m not sure about this but I believe that with a turbo engine more tumble isn’t always automatically better. Even if faster average burn would always be better, there’s also the variance of burn speeds that one needs to worry about, especially when knock limited on pump gas.

The piston guiding effects you speak of are related to high compression motors?
Read the Honda papers about tumble in a turbocharged motor. Super interesting and it matters a lot. More tumble means higher compression and detonation resistance which equals higher thermal efficiency and performance. They found like 2x more tumble in the NSX turbocharged engine vs the NA version of the same engine.

Honda does point out more tumble isn’t necessarily he most important, but the quality and aggressiveness (my word) is to mixture motion. That goes back to Logan’s post.

The piston guiding effects matter always. The way they design these ports is it hits the far side bore wall and runs down, across the top of the piston and back up.

They’re all well worth the effort of signing up (for free). They also have all their 2010 Formula One engine development stuff in there, which alone is super interesting.
Last edited by hoffman900 on Wed Sep 11, 2019 9:10 am, edited 1 time in total.
-Bob
hoffman900
HotPass
HotPass
Posts: 3445
Joined: Sat Feb 23, 2013 5:42 pm
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by hoffman900 »

See attached screen shots.
A593CA2A-F299-46D4-A922-9BB8257E999B.png
668EDBBF-0622-478B-84E1-2CE50E627A1E.png
You do not have the required permissions to view the files attached to this post.
-Bob
hoffman900
HotPass
HotPass
Posts: 3445
Joined: Sat Feb 23, 2013 5:42 pm
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by hoffman900 »

This one is great too. Gets into piston top shapes, squish shapes, and piston crevice volume, and their effects on mixture turbulence energy.
F6F92C47-5584-4046-80BD-BFF995C3037C.png
You do not have the required permissions to view the files attached to this post.
-Bob
sbcharlie
Pro
Pro
Posts: 220
Joined: Mon Jun 06, 2011 11:44 am
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by sbcharlie »

we race this engine in AMA flat track .started working on this engine in 2010 when they came out. did quite well we got the number 1 plate in 2013 with Wyatt Mcquire riding my bike. per AMA rules we can not change valve diameter. we did use 50 degree contact seat angles. on the head we build now we use what Honda has developed. we machine out the existing valve seat and than offset bore 1mm works quite well. if you get chance look at a crf Honda 450 cylinder head sbc
LoganD
Pro
Pro
Posts: 227
Joined: Wed Feb 20, 2019 10:13 am
Location: Rochester Hills, MI

Re: Theory of 4 valve poly quad cylinder heads.

Post by LoganD »

Modern turbocharged engines run a huge amount of tumble. Not only does this help with mixture uniformity, it actively cools edges in the cylinder and prevents hot spots. It's like....having a fan on the inside of the cylinder before combustion starts.

This is true even for port injected engines, probably the best production example is the Ricardo designed McLaren V8. Those heads rate very high on the tumble index. The reason you don't see DI on superbikes is that they need all the valve area they can get and it's not possible to package central DI with their valve sizes and cooling jackets. As was stated earlier, superbike engines have quite a bit of tumble, although not as much as say....the new Civic Type R engine or A45 AMG engine.

If you want to see radical cylinder head designs for swirl, look into the Supertruck project. There were some very interesting designs and they reached 50% thermal efficiency with a conventional turbodiesel.
hoffman900
HotPass
HotPass
Posts: 3445
Joined: Sat Feb 23, 2013 5:42 pm
Location:

Re: Theory of 4 valve poly quad cylinder heads.

Post by hoffman900 »

A lot of this is why I think performance on the 4 valve conversion heads suffer, just lack of knowledge of how these heads work.

Would love to see someone rip off a modern Sport bike head (like a Ducati twin, with a 4.5” bore) and apply it to a pushrod V8 shortblock. Obviously it would cost a fortune, but the head would be billet and the valvetrain would be the same design and all OTS.
-Bob
LoganD
Pro
Pro
Posts: 227
Joined: Wed Feb 20, 2019 10:13 am
Location: Rochester Hills, MI

Re: Theory of 4 valve poly quad cylinder heads.

Post by LoganD »

hoffman900 wrote: Wed Sep 11, 2019 9:29 am A lot of this is why I think performance on the 4 valve conversion heads suffer, just lack of knowledge of how these heads work.

Would love to see someone rip off a modern Sport bike head (like a Ducati twin, with a 4.5” bore) and apply it to a pushrod V8 shortblock. Obviously it would cost a fortune, but the head would be billet and the valvetrain would be the same design and all OTS.
I mean, bolt on Coyote's are making 500 whp on regular pump fuel. People are going 9's in Coyote stock (sealed factory stock engines). Making 700 hp from a Coyote is not hard. The aftermarket frequently makes two mistakes:

1. Deleting VVT, the Coyote's biggest advantage
2. Improperly degreeing aftermarket camshafts (a big mistake). You can't just install aftermarket cams and go with a DOHC design as your actual cam timing is dependent upon how they are installed.
Post Reply