Diy Porting

[mmt 15]

Thanks for all your replys guys.

 

I think me and Dremel will have a go sunday. At least all imperfections will be sanded down.

[petert 588]

Keep in mind that S16 (XU10J4) heads tend to have more metal under the seats (and less core shift) than Mi16(XU9J4) heads. As can be seen in Guy Crofts pics, it's possible to achieve reasonable bending of the seat into the port. Not so on an average XU9J4 head without installing bigger valves.

Edited November 22, 2009 by petert

[kyepan 291]

was reading david vizards how to tune a series engines last night, and can now understand a bit more about why it's hard to improve on these heads.

 

Am starting to understand a few things finally, and am mostly regurgitating what i read last night.. and i'm just checking i've got it right, so if you could respond and confirm please do.

 

3 angle seats - think small, think of the valve as a door, and you're just cracking the door open and the air has to get through the gap, but the door edges are all sharp, so you round the edges of the door frame to help. the point of this is to help air flow through the port past the valve at very low lift, instead of two big steps either side of the seat, when viewed in profile its a little venturi. However when the valve is on the seat it cools the valve, so you don't want too little material in contact, plus the impact needs area to absorb the shock, so there is a mechanical minimum too.

 

Inlet valve - so if this is the door you want air to release off the back of it easily as it fills the combustion chamber, like the back of a car.. so they design the back of inlet valves with sharp ish edges to allow a clean release of gas off the back of the valve into the combustion chamber.

 

exhaust valves are the opposite, as gas is flowing over the front of the valve and out of the port, vizard recommends that they are radiused as much as possible to help air get around the valve.

 

 

have not read any more than that...

 

Mini's have two types of head, small and big combustion chamber heads. The small chamber heads have a big lump in between the valves. which shrouds the inlet valve. Also they have lots of steps and lumps that can be smoothed out on the curve down to the port. Then the port shape can be much improved again by removing material.

 

Has anyone with a dead MI head chopped it in half to have a look at the actual port shape.. as that would be very interesting.

anyone with a dead head willing to do this for experiments sake.

 

Vizard also said that bigger valves don't actually help much unless they have accompanying work to remove restrictions elsewhere.

 

This got me thinking that the whole design of reciprocating engines with plunger style valves is stupid, because valves provide massive restriction to flow for most of the time as they are either closed or opening or closing... but only ever open fully for an instant. ...and am pretty shocked that no one has designed some kind of alternative.

 

other than a rotary... which uses ports like a two stroke.

Edited December 14, 2009 by kyepan

[DrSarty 90 1 Cars]

Great reply J.

 

I'm not saying no head ever requires 'work' or at the very least imperfection removal, but I gather from several sources that the Mi/S16 heads (and perhaps others, not just from PSA) are simply not worth touching, as the gain is either minimal or not possible to capitalise on without more extensive work (read: cost). This is why my reply and that from several others is as blunt as it appears to be.

 

But...head porting and polishing has quite a reputation, and therefore the perception is that it HAS to be a good thing to do, and therefore I understand why people feel it's something they SHOULD do. They then perhaps feel that if they don't, then they're missing some power they could've had, which I don't believe is strictly correct.

 

And to clarify the last point, if you do a YouTube search for head porting etc, you'll see some American heads for V8s, and the casting quality is atrocious! No wonder it's a big subject talked about a lot, and hence the porting reputation grows. But in this case it's just coping with bad design/workmanship, that the Mi16/S16 heads DO NOT suffer from out of the box.

 

So we're back to square one. They are excellent heads. Unless you're going to go stupid in your power hunt and are prepared to spend (a lot) to do it, they are really not worth touching. A little clean up to do with matching ports with inlet/exhaust joins and off you go.

[Tom Fenton 1,542]

Be careful comparing the siamese A series head to an Mi16 head. The A series definitely has a number of idiosyncracies (sp?) that set it apart from most other cylinder heads. Vizards is a very good book, but not all of it applies to other engines.

[DRTDVL 4 1 Cars]

Kyepan - also have a look at fluid flow in pipes and the boundary layer principal as these relate to how the air (fluid) will flow in the port (pipe).

 

Most home / DIY people when they polish the port think it needs to be a mirror surface, this is very wrong... Typically you will "polish" your ports with a 120-150 grit sandpaper hard roll, some people will recommend you only go as smooth as 100 grit. Anything smoother and you increase the boundary layer against the port wall.

 

Some factory performance heads are completely shocking, the Misty Mirage/Colt Mivec head (think their vtec) that we did some work on was absolutely shocking, in some of the chambers the valve set to port wall alignment had a 3-4mm mis-alignment, it would have create massive turbulence over the seat and in the port killing the flow, there was similar steps on the intake manifold/intake port transition. Flow bench figures are very very very very easy to fudge, and realistically should be looked at the same as a dyno... they are a tuning tool that is only comparable to itself, no point measuring cfm from different flow benches. I've seen 10cfm variances on the same head (untouched) by the operator playing with the putty around the port entrances, and similar variances by using mock intake runners. Inches of water will also effect figures, 10, 25, and 28 inches are all fairly standard with 28 being more common when i was working for my old company.

 

Depending on the quality of the surface finish some gains can be gotten by just polishing the port. We would typically see a 4-8% increase in port performance by polishing a factory as cast head.

[kyepan 291]

Kyepan - also have a look at fluid flow in pipes and the boundary layer principal as these relate to how the air (fluid) will flow in the port (pipe).

 

Most home / DIY people when they polish the port think it needs to be a mirror surface, this is very wrong... Typically you will "polish" your ports with a 120-150 grit sandpaper hard roll, some people will recommend you only go as smooth as 100 grit. Anything smoother and you increase the boundary layer against the port wall.

 

Some factory performance heads are completely shocking, the Misty Mirage/Colt Mivec head (think their vtec) that we did some work on was absolutely shocking, in some of the chambers the valve set to port wall alignment had a 3-4mm mis-alignment, it would have create massive turbulence over the seat and in the port killing the flow, there was similar steps on the intake manifold/intake port transition. Flow bench figures are very very very very easy to fudge, and realistically should be looked at the same as a dyno... they are a tuning tool that is only comparable to itself, no point measuring cfm from different flow benches. I've seen 10cfm variances on the same head (untouched) by the operator playing with the putty around the port entrances, and similar variances by using mock intake runners. Inches of water will also effect figures, 10, 25, and 28 inches are all fairly standard with 28 being more common when i was working for my old company.

 

Depending on the quality of the surface finish some gains can be gotten by just polishing the port. We would typically see a 4-8% increase in port performance by polishing a factory as cast head.

 

Cheers for the reply.

 

am i correct in thinking that the boundary layer is the air closest to the wall, which is attached to the wall either turbulent and slow moving or not moving, and the bit beyond the boundary towards the centre of the pipe is in laminar flow and freely moving.

 

valve set to port wall alignment - could you give a practical explanation of how this can be misaligned... does it mean then port comes down in one place and then the valve is offset?

 

J

[DRTDVL 4 1 Cars]

Overhang and underhang is where i'm refering to the relation of the seat in the port.... for example (this is hard to do with out a picture):

 

 

....\

......\ <--- seat

...|

...| <---- port wall... i would call this overhang

 

 

..\

....\ <---- seat

.....|

.....| <--- port wall i would say well blended

 

 

\

..\ <--- seat

......|

......| <---- port wall underhang...

 

 

 

Your effectively trying to get the biggest area of undisturbed air going through the port, while keeping the port velocity in thought also... So many people hack the crap out of the port chasing massive cfm figures but forget their cam's will not allow them to reach the lift that is required to make those figures. If you getting your head professionally ported the person porting the head, should ask you for the specs on the cams that you will be using so he knows the lift to work the port to... No sense making a port flow 10,000,000cfm at 14mm lift when your cam's only lift to 10mm at which the port has no velocity and thus you have no power... Also always ask for your flow figures done with and without valves, without valves will typically give you larger flow numbers for bragging at the pub and with valves@max lift of your cams will give you what you head will actually do - ideally you will get a range of flow figures at stepped ranges along the lift points of your cams...

Edited December 14, 2009 by DRTDVL

[DrSarty 90 1 Cars]

Fascinating!

 

So glad the OP posted now.

 

Healthy debate on this forum - when done well - is pretty amazing.