AGU Head

[Ibiza sport man]

The throat of the seat is the flow restriction within the port and it sets the desired port airspeed in relation to the bore size of the engine. By opening up the throat you will slow the port velocity sacrificing low lift gain in favour of high lift gains. As the cams for the 1.8t's are all considered low lift (less than 440thou) it is pointless to unnecessarily open the throats up as the restriction favours driveability.

I agree with you about the throat size and desired airspeed etc. What I find strange is the the port immediately behind the seat is so much larger than the throat with such a sharp step. Any good reason for this ?

 

[Backdraft]

Just a little food for thought in order to help clarify Hooch's comment:-

Traditionally 99% of multivalve intake ports that you see have knife edged port divisions. The theory is the thinner the partition the easier that air passes over it.

The reality of it is air will pass over a knife edge like a radiused one but as it does it tumbles and rolls of the edge. The resultant effect means that the incoming air passing over it has to first overcome the turbulance around the area of the port division before entering the combustion chamber.

In a real life situation this means that you have a slower throttle response in respect to a port division that is nicely radiused.

For those that don't understand the science behind it, it's around you everyday of the year in the form of an airplane wing. Get the radius correct and you can get air to stick to the surface instead of lift away and get turbulance. It's called the coander effect. Anyone that dissagrees please show me an airplane wing with a knife edged leading edge.

As for porting small port vs standard large port. A ported small port gives better throttle response and a flatter less spikey power curve to that of a large port head due to improved port velocities. The result is a much more enjoyable road car. One of the reasons why VAG introduced the small port head for their 225hp road going cars instead of fitting a largeport head as found on the earlier 150hp models.

A correctly ported small port head will comfortably support 500hp on a GT3071R with 0.82 backhousing as proven by Andy with his 10.2 sec white MK1 golf. Compared to a standard big port head that runs out of puff at 450-460hp on the same turbo and boost settings.

JP saved me having to explain everything!

 

[JNL]

I agree with you about the throat size and desired airspeed etc. What I find strange is the the port immediately behind the seat is so much larger than the throat with such a sharp step. Any good reason for this ?

It acts like an air reservoir. By making the restriction at the last possible minute you have maximum pressure (in the case of turbo) sitting behind the valve as it opens allowing for the maximum amount of air to flow through the valve during the time it's open.

Make the port smaller and you are now not feeding as much air to the open valve, it's a common practive on multi valve intakes that utilise low lift cams. If you want to take it to a real extreme have a look at the seats on the Suzuki Swift heads the old 1.3 ones that is (G13B). The seat starts to radius out and then has an a ledge encroaching on it reducing the throat size going straight into the 45degree seat. So your throat restriction is now a perfect 45 degree knife edge.

It is one of the few factory heads that have picture perfect even port velocity throughout the whole length and cross sectional area of the port in standard form. They are also reknowned for having an incredible amount of tq for such a small cc engine compared to other marques offering similar cc.

 

[Ibiza sport man]

For those that don't understand the science behind it, it's around you everyday of the year in the form of an airplane wing. Get the radius correct and you can get air to stick to the surface instead of lift away and get turbulance. It's called the coander effect. Anyone that dissagrees please show me an airplane wing with a knife edged leading edge.

Hmmm. There are loads of highly technical articles about the Coanda effect but this sums it up nicely.

Fallacious Model of Lift Production: You may have heard stories saying that the Coanda effect explains how a wing works. Alas, these are just fairy tales. They are worse than useless. http://www.av8n.com/how/htm/spins.html#sec-coanda-fallacy​

Surely it's Bernoulli's Law that applies to flows in ports ?

 

[Ibiza sport man]

It acts like an air reservoir. By making the restriction at the last possible minute you have maximum pressure (in the case of turbo) sitting behind the valve as it opens allowing for the maximum amount of air to flow through the valve during the time it's open.

So you're saying that the few cc of air sitting in this little chamber just behind the valve has more effect than the several litres in the port and manifold ?

 

[Feel]

So you're saying that the few cc of air sitting in this little chamber just behind the valve has more effect than the several litres in the port and manifold ?

He says air reservoir, but I think what he's referring to would be the effect of pressure rather than volume. Higher pressure just behind the valve would help, although whether this works in boosted applications...

 

[JNL]

Bernoulli's Law shows the dynamics between moving fluid/gas through a given cross section/volume and comes into play very much so when setting the airspeed of the port.

As for the Coander effect it shows the ability of air to not seperate from a surface if the curvature is correct. Whether they are using it correctly in the explanation of wing design or not is neither here nor there, that is where it is used most. My reference to a wing is if you knife edge the front of it the plane will struggle to fly due to imposed turblance from the leading knife edge.

If I take a head and place it on a flow bench with a knife edged division it screams so loud when I started pulling a vacuum it hurts one's ears. Utilising a pitot tube to measure velocity you can then see the port velocities are all over the place ranging from 100fps through to 254fps @10" when checking them at different points of the cross section, the ideal being between 185-209fps @10". More importantly when placing the pitot tube directly within 1-2mm and against the sidewall of the port division you can get 0fps velocity readings. Reverse the polarity of the probe so you are measuring what is coming out of the port instead of going in and you will get 180fps of air moving the wrong way. This shows there is air tumbling there. Take a head with a radiused port division it will not scream on the flow bench and you can quite happily test without ear defenders if you can bear the sound of 8 vacuum motors, the velocity readings now become more stable across the cross section of port and more importantly are showing that air is now all flowing in the same direction into the back of the valves.

As for the other question, air doesn't like going round corners, so by giving volume in the bowl below the valve, (mathematically the diameter of the bowl should = the diameter of the valve) you are giving it a place to accumulate to aid the change of direction for when the valve opens. Doesn't matter how good the rest of the port intake mani boost run etc.. flows if you can't make the final turn then it all goes to waste. A flow bench can't simulate this as it only shows constant flow at a given lift point, so this information has been worked out from years of personal experience and communication with people who have been at the top of their game for many more as we are always learning.
One of the most enlightening people I have ever come into contact with has been porting multi valve motorbike heads for 27yrs now his exact words where don't worry about cfm gains on multi valve intake ports shoot for an even velocity across the port. That statement changed my outlook once I saw the real life results. Several years down the line, after having explained the significance of port velocities to a customer the other week; they forwarded me this link after they did some more personal research to understand what I was on about enginebuildermag.com/Article/1482/drag_racing_cylinder_head_selection.aspx (sorry you'll have to copy and paste as I need to have made 15 or more posts appafently due to house rules) Although it is refering to single valve NA application it makes for very enlightening reading.

As for increasing plenumb chamber volume and rectifying runner dimensions etc it is very important especially the further you go up the rpm range as the plenumb acts as a reservoir to keep feeding the bowl below the valves everytime they close on FI application much like running a 150ltr tank on your compressor over 50ltr tank, the runners ensuring the smoothest transition between the two points.

Remember an engine is only as efficient as it's combination of components no point trying to shove more in if you can't get it out either.

Prime example is my mates Procharged small block chevy in his Porsche 924. Set up the engine on the dyno with the exh manifolds only, made just under 600hp and just over 500ft/lb using 2" collectors. Bolted into his car with a single 2.5" system and he only made 390hp and 320ft/lb. Obviously annoyed at his rather expensive toy I told him to hammer the floor out to make clearance for 2.5" collectors and a 4" system. With 4 people sitting in the boot on drag radial smoking up on the rollers it clipped just over 700 ft/lb and 680hp below 5000rpm as it started to jump out at higher rpm, just by getting rid of the bottleneck in the system.

 

[Indy]

Hi JP, refering to the last paragraph about the Porsche 924, looking at my setup currently I'm running a 3" down pipe - 2.5" middle section and back box. If I was go 3" all the way will that give me more ft/lb torque and possible add afew more horse power?

 

[Ibiza sport man]

Some interesting stuff in that article you refer to. But to go back to what I wrote that started off this discussion.

I must say the bigport head as standard has some pretty poor matching of the inlet throat to the back of the valve seat. Must be a 3 mm difference in diameter presented to the airflow as sharp step. It's like they've overbored it just behind the seat land. You can't improve it much and still retain a 3 angle seat.​

What I can't understand is why they have made such a sudden change in section just behind the valve seat. Is it to introduce turbulence and tumble into the flow ?

 

[JNL]

Like said before maximising airflow to the back of the valve and then controling the airspeed of the port with the seat to bring it in line with the cam profile. If you open up the seat which you can do whilst still retaining the important two angles of the seat the airspeed goes silly and hits almost 350fps in places on the standard port dimensions. You will get no more cfm out of the head because it started choking back at 209fps infact we recorded almost 5% less at 450thou lift.

If you start modifying the port dimensions to slow it down again you will get back to the same amount of flow as you do by just having kept the standard valves. The only difference is you end up with a very stable even port speed of 185-190fps pretty much anywhere you measure across the cross section and depth of the port.

This stability will give improved throttle response and more importantly allow you to turn up the boost a lot further before the head starts to choke compared to the standard unported big port head which varies from 190-245fps with a couple of zero's thrown in for good measure.

There is a tumble build into the port design but that comes from the port shape rather than the seat configuration which aids fuel efficiency by improving air/fuel mixture as it enters the combustion chamber.

 

[JNL]

Hi JP, refering to the last paragraph about the Porsche 924, looking at my setup currently I'm running a 3" down pipe - 2.5" middle section and back box. If I was go 3" all the way will that give me more ft/lb torque and possible add afew more horse power?

Not without a remap to suit and it wont be as significant on your car, the Porka was an extreme example that we happened to have the pleasure of experiencing because he had an exhaust system originally setup for a naturally aspirated 3.5ltr rover. Also ask Hooch how loud his was not the most pleasant thing to drive on the road noise wise especially if you think yours is loud already

 

[h17och]

loud is one word. but i think it suited the style of car. ie stripped out slag

indy, if yours is a daily id say its a no no. i ran full 3" with nothing but a backbox

 

[David_smith_909]

loud is one word. but i think it suited the style of car. ie stripped out slag

indy, if yours is a daily id say its a no no. i ran full 3" with nothing but a backbox

that vid never fails to make me think to myself "f**kin hell thats quick"
then i start chuckling at your passanger

 

[Indy]

Hey Dareen, that is supper loud. Looking at the video it looks like you need more than 2 hands on on the wheel cause of the torque steer. Mine is not going to be a daily car infact I'm looking for a bolt on roll cage for the back half.
My exhaust is quite loud as it is and from the speaking to JP regarding the EGT temperature, JP mentioned that if I ran a full 3" exhaust system than that would help off set some of the temp.

loud is one word. but i think it suited the style of car. ie stripped out slag

indy, if yours is a daily id say its a no no. i ran full 3" with nothing but a backbox

 

[ibizacupra]

egt's did'nt climb much on yours indy
low 800's to mid 800's

got that handling sorted yet?

 

[Indy]

The handling is a top secret should not be discussed
I've recieved a pair of complete track rods so will be changing those this weekend.

Bill the EGT reads around 875 when crusing on 90mph.

 

[ibizacupra]

remember that wire is busted indy on that thermocouple from being twisted on itself during installation by whoever.

it was'nt running at those levels here
not sure I would trust the figures where there's a known broken wire in that thermocouple

'secret suspension' - lol - get it off asap before it sticks you into a hedge