Inlet Port Flow V's Port Wall Finish

[3.5cologne]

I'm interested in hearing from experienced head ports on here in regards to inlet port flow v's port wall finish.
I have recently finished my inlet port rework on my B/V 5 port Cologne V6 heads and in the process learnt that there is quite a measurable difference in flow in relation to port wall finish.
I had read that polishing a inlet port didn't help port flow but with my newly built DIY flowbench last year, it's been a bit of a eye opener as to how much the port finish effects the flow.
After finishing the porting of the inlet ports and then comparing my porting work to images of professionally ported heads on the internet. My port work straight from grinding wheel or die cutter finish looked untidy in comparison.
So I decided to tidy the port finish up with a very course emery-cloth flap tool, leaving what I thought was an improvement to looks and flow. Hardly what I would call a polished finish at all.
Well that was enough to lose 7cfm, which in regards to Cologne heads is quite a significant loss in flow. I pulled the head off the flowbench and lightly ran a grinding wheel over the port walls to rough it back up and tested it again. Flow restored.
So do the experienced head porters on here find the same results ie: anything smoother than straight off the grinding wheel/rotary burr will lose flow, or am I missing something?
I'm just a weekend porter that works solely on my own heads, so keen to hear and learn what the experts thoughts are on the subject.

This port finish lost 7cfm.

[Graham]

hard to say, because i never finish anything with a finer finish than you have, and if you look at all the latest crop of cnc heads none of them have a smooth finish

[katana]

Google 'Boundary Layer Flow Separation'

[3.5cologne]

Google 'Boundary Layer Flow Separation'

Thanks Katana. That's what I was after, a cause or a reason for the loss of flow and something to look into.

hard to say, because i never finish anything with a finer finish than you have, and if you look at all the latest crop of cnc heads none of them have a smooth finish

Yes the CNC heads definitely have the machining marks left in them along with the golf ball finish type ports, but there is still a truck load of photos on the net of professionally ported heads with very smooth ports. I suppose I was quite surprised at the amount of lost flow it caused.

[Dyno]

I'm currently helping my friend to finish his 2600 Cologne turbo engine. I've done a small porting / seat bending job. What's the kind of CFM you could get out of you ported heads? In standard form I don't think they flow very well and even ported I believe it's not so wonderful.

The reason we decided the turbo instead of trying to port an almost impossible head.

Thanks

Dirk

[Graham]

The reason we decided the turbo instead of trying to port an almost impossible head.

Thanks

Dirk

makes perfect sense

[Dyno]

And exact the same as what MAY did. Probably frustrated trying to make any power out of this engine.

[3.5cologne]

I'm currently helping my friend to finish his 2600 Cologne turbo engine. I've done a small porting / seat bending job. What's the kind of CFM you could get out of you ported heads? In standard form I don't think they flow very well and even ported I believe it's not so wonderful.

The reason we decided the turbo instead of trying to port an almost impossible head.

Thanks

Dirk

I think the 2.6 Cologne heads are slightly different than the 2.8 heads. Not sure how they compare, I'm lead to believe they have smaller ports than the 2.8.

My 2.8 heads were already ported and fitted with big valves by a head porter here in NZ. I just reworked the inlets ports, so I don't really have any data on the exhaust side of things yet.

Anyway here's my flow data for the inlets.

44.3mm Inlet Valve CFM @10" (tested @28")


VALVE LIFT__Average Rework__Before Rework__Standard Head

0.20_____71.8_____72.2_____66.6
0.25_____87.0_____86.4_____78.4
0.30_____99.3_____94.0_____84.4
0.35_____105.0____97.4_____85.2
0.40_____109.3____98.7_____87.5
0.45_____109.0____98.9_____87.5

[Dyno]

A nice increase, difficult to compare with number taken at 28" but further more, it also illustrate how difficult it is to improve these ports.

Yes, 2,8 Ltr have larger ports, larger exhaust valves as well. The customer already started with the 2,6 ltr head before I came into the project. I would also have started from 2,8 ltr (even complete 2,8 Ltr engine).

I believe 2,9 Ltr head (and block) is still better but this was turbo orient from the first day, I don't care too much.

[Graham]

inlet flow figures are interesting, because with those inlet flow numbers you could see getting on 300bhp so clearly something else is holding these engines back badly,

[Miniliteman]

inlet flow figures are interesting, because with those inlet flow numbers you could see getting on 300bhp so clearly something else is holding these engines back badly,

Don't these heads have siamesed ports?

[Graham]

Don't these heads have siamesed ports?

siamesed exhaust ports, and if its a k jet 2.8 injection it will have a Siamesed intake manifold too

[Graham]

A nice increase, difficult to compare with number taken at 28"
.

it doesnt make any difference, well certainly not at the airflow rates we are talking about here, drag race v8s with huge aftermarket v8's might be a different matter, but in our world you get the same number whether you test at 10" or test at 28 and convert to 10"

i even tested some heads @5" converted it to 10", then retested at 10", later on retested the head on my bench @28, in all cases we got the same end result

[Dyno]

it doesnt make any difference, well certainly not at the airflow rates we are talking about here, drag race v8s with huge aftermarket v8's might be a different matter, but in our world you get the same number whether you test at 10" or test at 28 and convert to 10"

i even tested some heads @5" converted it to 10", then retested at 10", later on retested the head on my bench @28, in all cases we got the same end result

I just don't have the converting number to go from 10" to 28". I do know the test pressure will not matter too much but sometimes you need to compare.

[Graham]

I just don't have the converting number to go from 10" to 28". I do know the test pressure will not matter too much but sometimes you need to compare.

flow numbers in original post were at 10" only the test pressure was 28"

to go from 28" to 10" multiply by 0.6
to go from 10" to 28" multiply by 1.67

[Dyno]

It should indeed be possible to get 270 Bhp out of this engine. It will be hard to see 200

[Graham]

It should indeed be possible to get 270 Bhp out of this engine. It will be hard to see 200

Siamesed exhaust ports and intake are a killer power wise!

[Dyno]

Yes indeed but I believe the 2,9 heads ain't that much better

[Graham]

im told 2.9 makes a lot more torque, ford de-sieamesed the exhaust ports, but i dont think they did anything to improve the inlet, by the time the 2.9 arrived the ybt was well established and we soon after got the 24v colongne so i suppose development just stopped as both the other engines made 200bhp as standard

[3.5cologne]

Siamesed exhaust ports and intake are a killer power wise!

Dead right Graham. Throw the standard K-Jet manifold onto the 100+cfm head and the flow to the valves is down around the 80cfm mark.
I could have got more flow from my heads, but what was the point when the inlet manifold just strangles it? I will be spending more time on my EFI manifold before it goes on the engine. Currently my EFI manifold
gives me around the 100cfm at the valve but I think I can improve on that.


The 2.8 inlets flow better than the early 2.9 heads and on par with the later 2.9 heads, but the 2.9 does get it's extra torque from the proper 3 port exhausts and a much better injection system.

The customer already started with the 2,6 ltr head before I came into the project. I would also have started from 2,8 ltr (even complete 2,8 Ltr engine).

Yes the 2.8 engine would have been better to start with especially when turbocharging as it has a stronger block.

[Graham]

Dead right Graham. Throw the standard K-Jet manifold onto the 100+cfm head and the flow to the valves is down around the 80cfm mark.
I could have got more flow from my heads, but what was the point when the inlet manifold just strangles it?

i sort of agree, in my experience it doesnt matter how good the inlet system is you will always see a flow reduction with it fitted, so you might find if you can pick up an extra 10 cfm on the bare head will still gain flow with the intake on, how much will depend on just how restrictive the intake is, but if you can only find another 1 or 2 cfm on the head it probably isnt worth the effort, that said any flow gain is worth more with a 6 cylinder over a 4!

[Dyno]

The turbo will only make 0,5 bar. We are talking about low level power / pressure. But indeed, I would have started 2,8 Ltr from the beginning.

BTW, these blocks are a lot stronger? I have no experiences at all with these engine.

In US they had 6 port heads from the very early Mustang V6. But all with smaller valves. I've never known why they did not used the same heads over here.

There used the be a company called "Swaymar" or something. They also had 6 port heads but probably US castings?

Why not fitting 3x DCNF on this engine. Will give a lot more torque and more CFM compared to you choking EFI manifold.

[Graham]

swaymar did do 6 port heads, but they were fabricated, theres no water where the head bolt passes through the siamesed exhaust port, so you could machine away that part of the casting and replace it with a block of material with two ports bored in it, and the head bolt would help hold it in place

[Graham]

iirc us 6 port heads had something different about the valve orientation so needed a different cam, which i think as also different in that it was chain drive so ran in the opposite direction, or put it another way the US 2.8 was much more like the 2.9

[3.5cologne]

The turbo will only make 0,5 bar. We are talking about low level power / pressure. But indeed, I would have started 2,8 Ltr from the beginning.

BTW, these blocks are a lot stronger? I have no experiences at all with these engine.

In US they had 6 port heads from the very early Mustang V6. But all with smaller valves. I've never known why they did not used the same heads over here.

There used the be a company called "Swaymar" or something. They also had 6 port heads but probably US castings?

Why not fitting 3x DCNF on this engine. Will give a lot more torque and more CFM compared to you choking EFI manifold.

Yes I believe the 2.8 blocks are a bit stronger for sure, the power/rpm level your going for I don't think it would be a problem.

I don't know either why Ford never used the American 6 port heads in Europe. They do flow even worse than the 5 port heads but have more potential than the 5 porters.
They do have different inlet port spacings to the Euro head so the inlet manifolds are not interchangeable.

Fitting 3 x DCNF's would be nice but I think the 6 port EFI inlet manifold, with it's longer port runners, should give better lower end torque and that's what I'm after, as it's a street only engine.
Plus the expense of the 3 x DCNF's when I already have the EFI manifold and ECU.

No Swaymar used 5 port heads and machined a block to un-saimese the heads just like the Saab V4's did.

Swaymar head

[Dyno]

OK, never knew they did it that way. Clever done.

Nothing wil give more torque as 1 throttle plate pro inlet. It is a huge different compared to any single (twin) throttle plate EFI or single carb setup. It the first step for real power and torque. And from this point, you can use a lot more cam duration.

[3.5cologne]

iirc us 6 port heads had something different about the valve orientation so needed a different cam, which i think as also different in that it was chain drive so ran in the opposite direction, or put it another way the US 2.8 was much more like the 2.9

The 6 port 2.8 US heads had the same valve orientation as the euro 2.8. It was the centre inlet ports that are different. They used the same cam gear drive as the euro engine.

[Dan Johnston]

Back in the 90's when i use to run a dyno we brought off Turbo Technics there red 2.8 single turbo 4x4 sierra apparently it was on a Motorshow stand at some point , that was 276 hp and 340 ft at a bar of boost it was very rapid esp when the turbo come in hard ... never like av gas thou lol

[Dyno]

Some pictures of the head I made ready for the V6 turbo. I've done a small clean up / reshape.

[Graham]

looks like a useful improvement Dirk

[Dyno]

I hope it is. My Flow Bench is still not finished (not even started after I bought all hard and software). So I can not compare before and after. But I do know it should be better as standard and I feel the minimum you should do with any head build (when you are @ this forum).

Sometimes I call this head "BLENDING". Take out all sharp edges, blent all nice into the valve seat and do the same after the seat. Repeat for valves. Try to think how air would like to pass the port and make it as easy as possible by giving it no abrupt turns. The energy needed is 20% for 80% improvement ever able to find. If you want the next 20% you need 80% of energy, so 4x more as what you have already done for only a marginal improvement compared to what you have done already. That's why real race heads are so expensive.

I also believe, the best port is the smallest port which give you max flow, or should I see max flow in relation to the size. Which means not always max possible flow, you need to keep in mind the relation between both and it also means, this is unwritten text (grey zone). Small ports give a lot of "pulse " tuning. One of the most important parts to make real power (together with head flow of course). See it as the 50DCO with lot of CFM is not the best carb for your 1000 cc Pre-Crosflow.

Since about 30 years I don't polish ports anymore, leave the ugly as grind (not many people feel it looks fast). There are 2 reasons for it. I also believe rough ports flow better and less fuel can stay condensed into the inlet track. 30 years ago we polished like mirror (2 stroke race engines) until a saw a real factory Italian FANTIC head. It was far from what we where doing and ports where at the same time also a lot smaller. The second reason why I don't polish ports it because I'm lazy from nature

[3.5cologne]

I hope it is. My Flow Bench is still not finished (not even started after I bought all hard and software). So I can not compare before and after. But I do know it should be better as standard and I feel the minimum you should do with any head build (when you are @ this forum).

Sometimes I call this head "BLENDING". Take out all sharp edges, blent all nice into the valve seat and do the same after the seat. Repeat for valves. Try to think how air would like to pass the port and make it as easy as possible by giving it no abrupt turns. The energy needed is 20% for 80% improvement ever able to find. If you want the next 20% you need 80% of energy, so 4x more as what you have already done for only a marginal improvement compared to what you have done already. That's why real race heads are so expensive.

I also believe, the best port is the smallest port which give you max flow, or should I see max flow in relation to the size. Which means not always max possible flow, you need to keep in mind the relation between both and it also means, this is unwritten text (grey zone). Small ports give a lot of "pulse " tuning. One of the most important parts to make real power (together with head flow of course). See it as the 50DCO with lot of CFM is not the best carb for your 1000 cc Pre-Crosflow.

Since about 30 years I don't polish ports anymore, leave the ugly as grind (not many people feel it looks fast). There are 2 reasons for it. I also believe rough ports flow better and less fuel can stay condensed into the inlet track. 30 years ago we polished like mirror (2 stroke race engines) until a saw a real factory Italian FANTIC head. It was far from what we where doing and ports where at the same time also a lot smaller. The second reason why I don't polish ports it because I'm lazy from nature

Thanks onyd for info, your "ugly as grind" looks good compared to mine Your right though it doesn't look fast and that was the reason for me to try and tidy up the one port to see if it was worth doing the rest of the ports.
After flowing the port at 109.9CFM, then after spending an hour smoothing and tiding up the port only to re-flow the head at 102.6CFM was a jaw dropper for sure. The good thing though, is it was a easy fix, just a light rough up with the grinding wheel. So that begs the question for someone who has just forked out good money for a professionally ported head and it turns up with nice smooth ports. They sure look good and fast but do they get the grinder out and rough up their brand new pride and joy?? I think not.
This was the reason I asked on here for others experiences on the subject just in case there was something I was missing. I had read for years people saying not to polish the ports but they don't say that even smoothing the port is just as bad as polishing the port. It has only been since I built the flowbench last year that I've got to see for myself just how detrimental to flow smoothing or polishing the port can be. But I suppose some head porters do it because if it looks good it'll sell and yet's be honest, they're just doing what most uninformed customers wants to see.


One thing I did find when porting these heads, is in the photo where I put the arrows, there is a kink as the port enters the valve bowl area. Reducing this kink always made a gain in flow. But be warned there is water behind that port wall.
Also directly opposite on the other side port wall, I consistently lost flow when touching that area.(photo is before I did the rework)

[Dyno]

The first thing you need to do is the so called "blending". Without a nice turn from port to seat it is a no no. Once this is all done you can make ports slightly larger where you feel it can make an improvement. From this point, you need a flow bench. Taking out metal at the wrong place can make the head look fast but totally destroy them. This is when it all start to be expensive.

I usual stop at the point it is going to be expensive but I'm sure I've improved the head fair enough compared with the time and effort needed.

I see a lot of heads, silly seat grinding and the port going straight into the valve seat, no 60° cut, not small turn, blend. The port as big as the inner size of the seat. No matter what more you do to these heads, they are lost forever. Basic head tuning is all about the seat and the port / combustion chamber very close to the seat. And of course the valve shape.

[gid]

Shoot me, but nothing beats a good hand ported job. I prefer this above cnc marks in the ports. The key is the sustainable mixture in the inlet port. As what has been thought amongst
enthousiats in the past is a smooth mirror like finish. It does not work. Because it may result in puddles of gasoline dropping out. You have to keep turbulence in the ingoing mixture.
A coarse of #120 sandpaper finish is more then sufficient.

[Dyno]

Most CNC company's offer a hand finish for little extra money. They talk about 1 or 2 CFM improvement over bare CNC.

[3.5cologne]

Shoot me, but nothing beats a good hand ported job. I prefer this above cnc marks in the ports. The key is the sustainable mixture in the inlet port. As what has been thought amongst
enthousiats in the past is a smooth mirror like finish. It does not work. Because it may result in puddles of gasoline dropping out. You have to keep turbulence in the ingoing mixture.
A coarse of #120 sandpaper finish is more then sufficient.

I used #40 grit emery cloth to finish the port that lost 7cfm so would say #120 grit would be way to smooth. I found the finish straight off the grinding wheel or rotary burr was best, not pretty but definitely a measurable improvement over any form of smoothing.

Hey onyd how did your friends 2.6 engine go in the end?

[Dyno]

As grind looks awful but I think with power in mind perfect. You just can't sell them this way to a customer hahaha, the want shiny finish. Take this head to a tuning show and you will never ever sell a single bolt or nut !!!

I've fitted most of the engine but had to keep it on hold. Summer and too much wok to do. I can only work on it when I got a little time, just can charge him full price, he know he has to wait from time time. Hope we can finish it before New - Year. Currently working on the turbo carb.

[katana]

The number of times you read of guys wanting to get a 'port and polish' job doesn't help - just losing that phrase from the tuning bibles would help! One thing I don't often see with car heads is the use of epoxy fill to reduce overly large inlet ports as this is relatively common in very trick bike heads. As bike inlet tracts are so efficient (relatively) reducing the port dimensions to increase gas speed shows real results - the use of 4 valves sort of helps as well. A good example of why rough over smooth works is 'shark skin' Rub it one way and its smooth, rub it the other, its course and jagged - the roughness effectively 'detaches' water from skin reducing friction, improving speed.

If you want some 'thought provoking' bedtime reading try this - http://mototuneusa.com/super_sonic_nozzle.htm I have no connection with this guy and what he says / claims just don't sound true but there is an awful lot of feedback that tends to support his claims!

[gid]

depends on what you've got. In some cases you might want/need to open the ports up to gain flow, but its mainly port shape you're after. In my engine they are pretty large from factory
so i did not enlarge them even more, just shaped them erea to the valve seat, and portmatching to inletmanifold. Same on exhaust, where i did little to ( hopefully ) improve things.
I wish i had a flowbench to test it all. David Vizard book describes how to make a simple flowbench wich he got good results with, main goal to see differences, not particular cfm ratings.
That is what i think too, as long as i can messure differences, i dont particular care about the cfm readings.

Good discussion here. keep it on.

[Dyno]

A flow bench is a very useful tool, just as a dyno. But sometimes max flow means bigger ports and bigger ports mean less pulse tuning. And you do need pulse tuning. The best carb for you 1100 Fiat is not 55DCO (it is the best flowing). I always say, the best port is the one that flow reasonable numbers without being to large.

You just need to make compromises, using a dyno to find the very last Bhp (and already not torque at all 500 RPM below max power) is NOT going te mean a good engine.