I built a flow bench.

Its a very simple design. It only measures resistance to flow (and therefore does not produce flow figures), but is still very useful as a tuning aid for cylinder head development.

I cut a hole in the top of an old chest of drawers, and stuck some drain pipe innit.

Here you can see the drain pipe with tube going to the manometer. You can see a few drops of water where the vacuum cleaner tried to drink the manometer water.

Attach El' Cheap'o Hoover (from argos) as the vacuum source.

Make a mounting plate (not yet finished. The large drill bit is just about the right height to poke you in the eye)

Fill manometer up to the zero mark. The water (and my hands) has been died red.

Then stick on a cylinder head and measure the flow resistance.

I made up a simple bracket out of some sheet steel that has had a thread cut to allow a blot to screw into it. I can open and close the valves using a bolt (above).

The idea is to have a little play with this cylinder head (its from the old fiesta OHV engine. Sorry its not a polo head. I don't have one), and see if I can create any improvement in flow (by reducing resistance to flow).

Got the idea from here:

http://www.gofastnews.com/board/technical-...flow-bench.html

I had a little go at porting the cylinder head.

I just removed the casting marks, and blended the valve bowl.

Here are the results.

You can see an improvement of about 2 at all lifts.

You will also notice that flow levels off at about 6cm valve lift.

I don't know how high the valve lift was in the old fiesta, but I guess it was higher than 6mm.

I think the shape or port size is restricting total flow.

I'll have a go at modifying the port shape to increase the flow over the next week.

No idea about horse power potential.

My bench only shows flow resistence, and will not produce flow figures.

So a low number is good (little resistance, and therefore high air flow), and a high number is bad (High resistance or bad air flow). The graph above seems inverted but shows what you would expect with the type of system I am using.

A.G. Bell suggests a port size of 0.67 of valve size for a square port. So I'll have a go at making the port that size. I'll also reduce the radius of the short side turn and grind down the valve guide.

I want to keep the basic shape of the port. It would be easy to get good flow numbers by just making the ports huge. But that wouldn't give good results on a car.

I plan to leave the port floor as it is, and remove meterial from the port roof and one side of the port. I also want to turn the valve down a tad. I'll take some pictures and make some drawings for on here so you can see whats going on.

Just thought I'd share this, not sure of the responce I'll get.

Any comments would be appreciated.

Thats bloody brilliant! How very resourceful :-)

there was an article in PPC mag not long ago showing how to make a very similar thing too.

Looks like you managed to get that red water all over the place :lol:

It'd be interesting to compare values between standard heads and ones that have been worked on professionally too. If I'd have thought about it I could have compared my big valve head with a standard one I have before I put it onto the engine recently.

there was an article in PPC mag not long ago showing how to make a very similar thing too.

Looks like you managed to get that red water all over the place :lol:

It'd be interesting to compare values between standard heads and ones that have been worked on professionally too. If I'd have thought about it I could have compared my big valve head with a standard one I have before I put it onto the engine recently.

That red water was a complete pain. I tried pouring it in, which didn't work at all. So I ended up using a syringe.

If I did it again I would just submerge the tubing in some water and then add a small amount of dye to it.

I do have two heads. Both Ford CVH 1.6.

One has supposedly been worked, (and the ports do look the part) and one is a standard head.

Only problem is they have different combustion chamber shapes, so I'm not sure how good a test it would be.

Apart from that and the fact the standard head doesn't have any valves I should be able to do a direct comparison between the two. I'll Just use the valves form the modified head in the standard one.

I need to modify the bench slightly. It varies on how much vacuum it pulls at different times of the day.

I'm guessing this is something to do with the temperature, humidity, atmospheric pressure and the power supply changing slightly over the course of the day.

I think a few holes which can be plugged/opened so I can get the same amount of depression at the manometer at all times will allow me to make sure each and every test is under the same conditions (or as close as can be).

So I modified the mounting plate. Using old head bolts which can clamp the head to the plate, and sticky back foam rubber to improve the seal.

This worked so well that when I tested to see how much the bench was pulling, it drank the whole manometer before I could turn the Hoover off.

So like I said above I made some vent holes. I can use these to get a constant 50cm depression on the manometer no matter what the conditions. Its not pretty but it works.

Right, I want to make the port 0.80 times the size of the valve. I was going to make it 0.67 but thats about the size it is already.

The valve is 1.25" or about 31.6mm.

The area is 784.27mm^2 meaning the port area will have to be around 627mm^2.

The port measured 25mm tall by 20mm wide, giving an area of 500mm^2. I know I haven't taken into account the rounded corners. But I don't really care.

Okay, so to get to the desired size I need another 120ish mm^2 on the port area.

I decided to make the height 6mm taller and leave the width the same. I've got 3 other ports to play with, so I'm just seeing what works and what doesn't

This is where i need to be.

And after an hour or so I'm just getting there.

I also did the valve bowl.

From this:

To this:

There is a very nasty step below the valve seat (looking down at it), so I blended this out. I also tried to blend in the short side. I'll work on this more later.

I should have roughly finished this port tomorrow so I'll try and get some flow figures.

Looks very interesting. I have a head i removed yesterday that i was hoping to Port+Polish. I can lend you the head if you need it.... you're not too far away from me either!

Looks very interesting. I have a head i removed yesterday that i was hoping to Port+Polish. I can lend you the head if you need it.... you're not too far away from me either!

Thanks for the offer.

But I'm going to have to decline.

Making a mess of an old cylinder head is one thing, but someone elses that they want back!

Theres no reason you can't do what I'm doing.

A few links for you:

http://www.clubgti.com/forum/showthread.ph...ghlight=porting

http://www.gofastnews.com/board/technical-...strictions.html

http://www.gofastnews.com/board/technical-...g.html#post5343

http://www.gofastnews.com/board/technical-...rt-volumes.html

http://www.gofastnews.com/board/technical-...port-areas.html

http://www.gofastnews.com/board/technical-...of-porting.html

The only tools I'm using are a corded drill with flex extention, grinding stones, and a few burrs.

I am planning to get an air powered die grinder however.

A digital caliper and a spring caliper set are also useful and can be picked up for less than £15.

Damn!

Looks like I'll be getting that die grinder sooner than I thought.

To be fair it hadn't been used for over 30 years so I didn't expect it to last long.

However before that decided to break itself I did manage to clear about halfway down the inlet port. Not to the amount I would have liked, but a fair amount of material was removed.

You can see how much was removed from the roof of the port here:

And here (also note the genius that designed this head has the port swirl the mixture into the cylinder wall, and not toward the center of the cylinder. Numpty):

I got it on the flow bench (which is behaving nicely now) and took some figures:

The results were, well........I'm not overjoyed.

You can see the modified port flows less at low lift, and more at high lift than the unmodified port.

I think I've opened the roof up too much and I've reduced flow velocity too much, resulting in poor low lift flow.

I did compare modified port 2 with unmodified port 3. So its not completely scientific.

I also don't think I'm getting enough depression from one vacuum cleaner. I may have to get another one.

I think I'll work on the short side radius to try and get some flow back at low lift. Most of the flow seems to be on the port floor, especially at low lift so I imagine that concentrating on the short side turn into the valve bowl will improve flow at low lift the most.

I will continue to open up the port roof as see what happens to the flow figures. I reckon high lift flow will improve and low lift flow will reduce as I continue to open the port up.

Hmmm, any ideas anyone?

I think theres a lot of scope for improvement......

Hmmm... I cannot contribute technically here but this is excellent work. I shall be keeping an eye on your progress!

PoloD.

I'm sure you've all been waiting for the next installment

(maybe not, BUT I'll continue to post them).

You need the right tool for the job:

And what a difference it makes.

I can not describe how much quicker and easier it is to remove the metal with the proper tool.

However I'm still lacking the proper tools.

I need a long shank to be able to get to the middle of the port.

I took a lot of material out of the valve bowl on the long side turn.

I wanted to improve the radius by making the turn shallower, or less steep.

I also cut back the head to the gasket line. If you've read the other posts you'll remember I said the port directs the flow towards the cylinder wall (and not the center of the cylinder).

By de-shrouding the inlet valve I hoped to improve flow, I could probably be more aggressive in this area.

I left some material for compression towards the spark plug.

Here you can see the work that has been done on the port entry, and valve bowl.

Also notice the lack of work on the middle area of the port (where I can't reach).

Heres more of the same but at a slightly different angle

Also I cut back the valve to hopefully pick up some extra flow.

So what has this done for the flow results?

I'm much happier with these figures.

The work has brought the port back up to the flow level of a standard port at low lift,

and has improved flow at high lift even more than before.

Any comments anyone?

I'll be happy to hear/try an ideas anyone has.

Edited 14 April, 200917 yr by varnish

how did you do the valves?

on a lathe?

or a into a drill chuck?

either way, great engineering. nice results

Nice thread, an interesting read. Would be cool if you tried it out on a polo 1300 head, any chance?

Valves:

Yes. In a drill chuck. And why not?

Using an old angle grinder disc. If I were to use the valves in a working engine I would finish them off with some sand paper.

I would like to get my hands on an old polo head at some point. If I had a couple it would be better.

So if anyone has any scrap heads I would be more than willing to take them off their hands.

I assume all the early 1.3 8V heads same?

I've got some plans for an improved flow bench that will allow me to measure flow rates.

I'll have to speak to my local engineer and my friend who lectures in fluid dynamics but I'll probably have something working in the next few weeks.

I think this will be my last update.

Unless anyone has any bright ideas, or I think of one myself.

I decided to start on a new port.

With the other port I took material out of the roof to make the port taller.

This time I took material out of the port floor (which convention says you should stick clear of)

The port roof, and one of the port walls.

The picture shows the port upside down.

In the picture its the right hand side of the port I machined.

I wanted to make the port straighter to reduce the swirl affect.

Swirl is usually advantageous. But in this head design it directs the flow straight into the cylinder wall.

Which brings us to the next picture:

I was more aggressive when removing material from the combustion chamber this time.

So much so the inlet valve stands proud in the combustion chamber when fitted.

I wanted to de-shroud the valve as much as possible.

Results.

Even better than last time.

Better flow at low, medium, and high lift.

Just a little bit about swirl. I've been going on about it and I guess most of you won't know what it is.

(I know, but have you ever used MS paint with a broken mouse?)

The port on the left has good swirl.

The port swings the air at a nice angle away from the cylinder wall, where it continues to swirl around in a circle and reaches the spark plug. It aids mixture distribution in the cylinder, and homogenisation of the air fuel mixture.

This allows good combustion, a clean fast burn which needs less spark advance and produces more power.

The port on the right has bad swirl design.

The air is directed into the cylinder wall. It bounces off the wall and does not move towards the spark plug (well not in a uniform manner), distribute nicely, or have good homogenisation. This is the design of the head I'm working on has.

Edited 13 April, 200917 yr by varnish

There is a very nasty step below the valve seat (looking down at it), so I blended this out. I also tried to blend in the short side. I'll work on this more later.

I should have roughly finished this port tomorrow so I'll try and get some flow figures.

Hard to tell from the pictures, but the step there might be there in order to create a better swirl inside the combustion chamber, so although you improve flow on the bench, it might give you less hp on a dyno

You can see the step in the picture below. On the inlet port of the far right combustion chamber.

All of this work is (pretty much) useless without dyno results.

It would be easy to just hog the ports out completely and get huge flow figures. I was trying to avoid doing this.

But as you say, everything I did could just give good flow results and prove useless when fitted to an engine.

I'll never know in this case.

What I really need is a test engine.

Edited 14 April, 200917 yr by varnish

Ah, can see it now, I woould smooth that too.

Normally the ones I was refering is a bit more "wing" shaped

That red water was a complete pain. I tried pouring it in, which didn't work at all. So I ended up using a syringe.

If I did it again I would just submerge the tubing in some water and then add a small amount of dye to it.

Just stick 1 end in the red water and suck on the other end

I think I'll work on the short side radius to try and get some flow back at low lift. Most of the flow seems to be on the port floor, especially at low lift so I imagine that concentrating on the short side turn into the valve bowl will improve flow at low lift the most.

I will continue to open up the port roof as see what happens to the flow figures. I reckon high lift flow will improve and low lift flow will reduce as I continue to open the port up.

Hmmm, any ideas anyone?

I think theres a lot of scope for improvement......

yea if it was me i would smooth out the valve bowl round the valve seat and work the short side turn, i wouldn't touch the port length, not unless it was to add in material to the floor. then you could possibly take some out the roof. thats alot of work though.

This time I took material out of the port floor (which convention says you should stick clear of)

The port roof, and one of the port walls.

removing material from the floor is never the way, it will only make the short side turn even steeper, if you add to the floor and take from the roof so the port is steeper, then the turn into the comb chamber is reduced. Problem is you will have to do the same to the inlet mani so it all matches, lot of work.

Removing material from the floor is never the way, it will only make the short side turn even steeper, if you add to the floor and take from the roof so the port is steeper, then the turn into the comb chamber is reduced. Problem is you will have to do the same to the inlet mani so it all matches, lot of work.

I know, but there was a step on the port floor. It seemed to drop down about halfway down the port so I opened it out to this point so the port was the same dept along its whole length. The short side radius remained the same.

I had seen this picture and wanted to try something like this:

I gave up when I saw how much work it would be.

I know about adding material to the port floor and removing material from the roof. In effect making the port more downdraft and making the entry angle to the back of the valve less steep. But like you say, a lot of work.

I felt I had to make the ports bigger as in standard form they are so small. I'll put up a picture of an unmodified port so you can see.

Anyway its just a technical exercise. I wanted to see what would work and what wouldn't.

What I seem to have achieved so far is a shape that has improved flow at all lift levels. Whether this would actually lead to an increase in power is unclear.

Edited 14 April, 200917 yr by varnish

It may be an idea to use modelling clay to raise the floor. Not suitable for use on a car but will allow you to easily shape the port whilst your playing with the flow bench.

From the pics it looks like there are no valve guides? If you have pressed them out then dont worry its too late. If not then think about removing the valve and then having a tickle until you get something like this. -

As for back-cutting the valves. Seen 8% gains over a standard G40 valve.

It may be an idea to use modelling clay to raise the floor. Not suitable for use on a car but will allow you to easily shape the port whilst your playing with the flow bench.

From the pics it looks like there are no valve guides? If you have pressed them out then dont worry its too late. If not then think about removing the valve and then having a tickle until you get something like this. -

As for back-cutting the valves. Seen 8% gains over a standard G40 valve.

The valve guides are cast in the head.

I was thining of using plasticine to see where the flow restrictions are most prevalent. Could also use it to see if removing more material will improve flow further. I might play with the head a bit more in the future.

Right now I've got an engine to rebuild and a car that needs painting. I will give that my first priority.

Edited 15 April, 200917 yr by varnish