Port observations on a Toyota V6.
Posted: September 5th, 2007, 8:51 am
Hello Guy, I hope this may be of some interest to both yourself and other forum readers.
The engine I am currently trying to improve upon is from Toyota's stable and is used in the 1992 to 1996 camry series of vehicles.
The engine designation is 3vz-FE, the "fe" part of the name denoting heads primarily for economy.
It displaces 3.0 litres with a bore of 87.5mm and stroke of 82mm, has 24 valves and 4 camshafts driven via a toothed cambelt on the intake cams only.
The exhaust cams are driven off a skew gear on the intakes and travel in reverse direction (anticlockwise), so phasing the cams for advance/retard could prove to be a little problematical later on.
My initial investigations into this head have been mostly by eye and finger and for the time being it will have to stay that way until I can begin to aquire the necessary airflow measuring tools and techniques.
The intake ports appear to be half decent judging by past experiences but suffer from sharp machined edges on the short side radius and various casting imperfections.
After pulling the valves, I found that these heads already have a multi angle cut to them on the intakes and a backcut to the valves.
The valves are inclined 11 degrees from the vertical.
Intakes have a 34.00mm head and exhausts measure at 27.35mm, so near enough 80% the size of the intake.
I havent yet measured/catalogued the seat widths and backcut angles, port volumes and throats etc etc.
I was interested to see how the air actually moved in these standard intake ports, so I made a dummy cylinder of the approximate bore size, sprayed the intakes with a white powder called developer (used in welding), and used some plasticine to put a doughnut radius on the intake port mouth.
I then set the inlet valves open an equal amount but within max lift for the cams and switched on the vacuum source.
I had already made a simple U tube manometer to give me comparative readings and I proceeded to move a probe down the port towards the valve.
As the probe approached the valve the reading on the manometer started to get higher which to me indicates the pressure is lower at that point.
The probe didnt tell me much initially and it is a little insensitive so I proceeded to the use of a dye sprayed into the airstream.
The amount of dye sprayed into the air was just the tiniest amount aerosoled across the port mouth and not into it.
After the vaccum was switched off the dye was clearly visible on the port wall but it had stained in only a small area, something I hadnt expected.
Initially I suspected the angle the head was sitting at was responsible in part for this odd staining due to gravity and so I set the head up in its normally mounted position and re tested with the dye but yet again it stained only the area you can see in the pictures.
At first I believed that the air was travelling moreso in the stained area rather than anywhere else, but I needed some other method to prove it to myself.
I aquired some smoke pellets from a local plumbers, a can with a screw lid and a short silicon tube and metal pipe to direct the smoke out of.
It worked rather well and the smoke was easily traceable as it traversed the port.
When directing the smoke to the stained area, some turbulence and instability was seen with the smoke billowing back and forth rather than travelling on that side of the guide boss also I was surprised by the back eddying that existed at the valvestem and guide interfaces. I suspect if the dye was substituted for fuel it may drop out of suspension in that area, but thats a pure hypothesis on my part.
The majority of air seems to want to flow to the outside of the ports approximately half way up the wall at its deepest point and round the short side radius, even though its quite badly formed.
From observing the smoke and dye staining it would appear that approximately a quarter, maybe even a third of the valve's circumference is not being utilised to full advantage.
I repeated the smoke and dye staining with different lifts but little change was observed in the position/route the dye and smoke appeared to take.
In an attempt to get to know these ports a little better and get some measurements i've resorted to moulding them with silicon.
The intake ports all have a near straight shot from the manifold but the exhausts come in two distinctly different shapes.
Intake.
Exhaust.
I have to say, I think that the exhausts look pretty dismal, so im hoping it should be possible to improve upon them by quite some margin, so my next port of call (no pun) is to cast a model in plaster to carry out some work to see where those improvements might lie on both intakes and exhausts rather than wreck a good head.
I was especially impressed by the smoke resolutely refusing to approach the port walls no matter how hard i tried to make it do so and the still air at the port walls appeared to be at least mm measurable.
I hope my rather amateurish attempts at understanding port flows have proven interesting, certainly for me it has confirmed airflows just dont go where you think they will and ive gained some useful insights.
The engine I am currently trying to improve upon is from Toyota's stable and is used in the 1992 to 1996 camry series of vehicles.
The engine designation is 3vz-FE, the "fe" part of the name denoting heads primarily for economy.
It displaces 3.0 litres with a bore of 87.5mm and stroke of 82mm, has 24 valves and 4 camshafts driven via a toothed cambelt on the intake cams only.
The exhaust cams are driven off a skew gear on the intakes and travel in reverse direction (anticlockwise), so phasing the cams for advance/retard could prove to be a little problematical later on.
My initial investigations into this head have been mostly by eye and finger and for the time being it will have to stay that way until I can begin to aquire the necessary airflow measuring tools and techniques.
The intake ports appear to be half decent judging by past experiences but suffer from sharp machined edges on the short side radius and various casting imperfections.
After pulling the valves, I found that these heads already have a multi angle cut to them on the intakes and a backcut to the valves.
The valves are inclined 11 degrees from the vertical.
Intakes have a 34.00mm head and exhausts measure at 27.35mm, so near enough 80% the size of the intake.
I havent yet measured/catalogued the seat widths and backcut angles, port volumes and throats etc etc.
I was interested to see how the air actually moved in these standard intake ports, so I made a dummy cylinder of the approximate bore size, sprayed the intakes with a white powder called developer (used in welding), and used some plasticine to put a doughnut radius on the intake port mouth.
I then set the inlet valves open an equal amount but within max lift for the cams and switched on the vacuum source.
I had already made a simple U tube manometer to give me comparative readings and I proceeded to move a probe down the port towards the valve.
As the probe approached the valve the reading on the manometer started to get higher which to me indicates the pressure is lower at that point.
The probe didnt tell me much initially and it is a little insensitive so I proceeded to the use of a dye sprayed into the airstream.
The amount of dye sprayed into the air was just the tiniest amount aerosoled across the port mouth and not into it.
After the vaccum was switched off the dye was clearly visible on the port wall but it had stained in only a small area, something I hadnt expected.
Initially I suspected the angle the head was sitting at was responsible in part for this odd staining due to gravity and so I set the head up in its normally mounted position and re tested with the dye but yet again it stained only the area you can see in the pictures.
At first I believed that the air was travelling moreso in the stained area rather than anywhere else, but I needed some other method to prove it to myself.
I aquired some smoke pellets from a local plumbers, a can with a screw lid and a short silicon tube and metal pipe to direct the smoke out of.
It worked rather well and the smoke was easily traceable as it traversed the port.
When directing the smoke to the stained area, some turbulence and instability was seen with the smoke billowing back and forth rather than travelling on that side of the guide boss also I was surprised by the back eddying that existed at the valvestem and guide interfaces. I suspect if the dye was substituted for fuel it may drop out of suspension in that area, but thats a pure hypothesis on my part.
The majority of air seems to want to flow to the outside of the ports approximately half way up the wall at its deepest point and round the short side radius, even though its quite badly formed.
From observing the smoke and dye staining it would appear that approximately a quarter, maybe even a third of the valve's circumference is not being utilised to full advantage.
I repeated the smoke and dye staining with different lifts but little change was observed in the position/route the dye and smoke appeared to take.
In an attempt to get to know these ports a little better and get some measurements i've resorted to moulding them with silicon.
The intake ports all have a near straight shot from the manifold but the exhausts come in two distinctly different shapes.
Intake.
Exhaust.
I have to say, I think that the exhausts look pretty dismal, so im hoping it should be possible to improve upon them by quite some margin, so my next port of call (no pun) is to cast a model in plaster to carry out some work to see where those improvements might lie on both intakes and exhausts rather than wreck a good head.
I was especially impressed by the smoke resolutely refusing to approach the port walls no matter how hard i tried to make it do so and the still air at the port walls appeared to be at least mm measurable.
I hope my rather amateurish attempts at understanding port flows have proven interesting, certainly for me it has confirmed airflows just dont go where you think they will and ive gained some useful insights.