Flow Benches and head porting

[GORDON STREETER]

Denis I would think that 190 bhp is a bit ott for a 1600 ! A good 1760 might get somewhere near it, and I can remember a Mass 2lt crossflow in an Escort
years back that had amazing hp . Going that big is a bit counter productive now as there isn't any historic class that it would go in .Also the cost to make one would be a lot more than a more modern powerplant. A well built 1700 is a more cost effective option as the bore size is not that much oversize, and if you start going 85mm or more the blocks are not that reliable.
My engine gives amazing hp and does rev to 10 000 but not a lot of torque, it is also very reliable . It also cost a LOT of money to build !

[Dennis.Doyle]

Gordon, I think we are in violent agreement on this one.

BTW I'm going "modern" this year with an early '90s 16V RS2000 - expect lots of silly questions on that one as soon as I've worked out what they are!

[phoenix]

Originally Posted by Notso Swift

That could be right if maximum flow occured only at maximum lift, but it dosen't. Generally the surface area is proprtional to area of the valve, and it translates pretty well to the capacity for flow. If the port is right I think (from memory here so don't cricify me if i have it wrong) maximum flow occurs from either 1/3 or 1/4 of the valve diameter. Generally, the steeper the ports (closer to the vertical of the valve) the earlier it can happen

But surely no-one would go to all the trouble and expense of 4 valves per cylinder for only a 3% advantage?

[zefarelly]

Mk IX 1963 1498cc 120-125 Bhp. Racing version of Mk VIII

my first race engine was roughly the same as this and produced 119 BHp at 6200 on the rolling road, it was 1500cc, twin cam valves and an A3 cam ( as this spec) it only had 10:1 CR though, I think a bit more would have produced an extra few horses at nearer 7k.

the 1500 prex wasn't officially developed much after 1963 so its down to Gordon, me and another few loons with hair and guitars!

[Notso Swift]

Originally Posted by phoenix

But surely no-one would go to all the trouble and expense of 4 valves per cylinder for only a 3% advantage?

That was comparing the standard valve size of a 4AGE to the modded crossflow, now I know Dennis has an old Atlantic spec motor I am thinking that they have +2.0 mm (30.5 mm) valves, so you are talking an extra 15% of air available.
The other thing is that 4v usually give a wider spread of torque, which is certainly desirable for a racecar, and essential for a road car. This extra air potential and spread of torque means you can run smaller cams for the same result, great for emissions – that is why the manufactures bother. I mean, Peugeot invent the 4v motor in about 1908, yet it was pretty much exclusively used for competition until the 1980’s, when the emissions laws started becoming tougher.

As for the 200 hp cross flows, I think you guys are on the money, they are made up of some inexplicably small and feeble horses! It is rare for a Pinto to get 200hp, leave alone a crossflow.

[Dennis.Doyle]

Came across some interesting looking ones today. OK, they do come from one of these companies selling the "200bhp crossflows" but bear with me - it's not the absolute figures that are interesting, it's the comparison of the 2 & 4 valve heads. For now, we'll just have to trust they didn't draw them in Powerpoint

The first is their big valve alloy xflow head:

http://www.cncheads.co.uk/ford_alloy_xflow_bv.html

The second is their Duratec 4 valve:

http://www.cncheads.co.uk/ford_duratec.html

To save confusion, just compare the very top lines, i.e. flows for the big-ported inlets.

At tiny lifts, say 50 thou, the Duratec is flowing twice as much as the xflow.
By max lift, ~500 thou, it's only flowing 50% more.

I think it illustrates nicely the difference of opinion between Pheonix & Notso. At the tiny lifts, Phoenix's view that the effective area for the gas to get in is a cylinder-shaped "curtain" defined by the valve, the port it's just lifted off & the curved surface joining them up. The differences between those areas on the 2 & 4 valve engines is absolutely massive at the low lifts. Conversely, when the valve is fully open, it hardly masks the port at all & it is the relative port areas (pretty much proportional to the valve areas) that matters. It is still a big difference, just nowhere near as big as the difference at the tiny lifts.

Does this make sense?

I think it's mostly of theoretical interest - except possibly when deciding what the important parameters are in choosing between different engines.

Your thoughts gents?

[GORDON STREETER]

Interesting, but after trying to cure a misfire problem before a Spa meeting last April the only proper test was on the track. The guys at Brands let me in on a track day at the last minute for a blast round . Putting the car back in the box some bloke came up and said "HOW MUCH FFF HP HAS THAT THING GOT" because I had pulled away from his amazing chipped scooby exiting clearways, and he had "at least 400 bhp" so I must have more!!

[greenamex2]

For me the most interesting thing about the figures is how little they managed to improve the flow on the inlet side on the Duratec (+17%) compared to the X-Flow (+60%). Clearly the manufacturers are getting pretty good.

It's also interesting to compare similar capacity engines rather than these links which is 1600 versus 2000/2300. Obviously the larger an engine the bigger port size it can 'tolerate', as I have found out to my cost!

The Puma head (1700) only makes 137.2 on the inlet side (+14%).

Another interesting point is how much better the the Duratec head flows than ANY of the other 2000 16 valve heads. It is even up on the Vauxall XE big valve by an appreciable amount. In fact the only one that beats it is the Coswort YBT, presumably because it starts off with such big ports and explains why they all end up having to rev so high.

[Dennis.Doyle]

Originally Posted by Denis Bassom

...
It's also interesting to compare similar capacity engines rather than these links which is 1600 versus 2000/2300. Obviously the larger an engine the bigger port size it can 'tolerate', as I have found out to my cost!

The Puma head (1700) only makes 137.2 on the inlet side (+14%).

Agree it was a very poor choice on my part for comparison. In particular I'd like to stop concentrating on the peak flows. Using the Puma/xflow example the same effect of much better flow at low lift is again evident. Difficult to read off the graph but maybe the difference between 16CFM and 23 CFM at 50 thou i.e. closer to 40% imrpovement with 4 valves at that point? While the flows at low lifts may not contribute that much to cylinder filling I can't help wondering how much assistance it gives the cam designers.

As I hope is clear, I'm just trying to get my head around what the significant differences are & hopefully become a more sophisticated consumer when it comes to buying things like big-valve/flowed 16V heads for the first time.

[greenamex2]

Something the American's do is also to list the volume of the ports, which gives you a very good idea of how 'efficient' the porting is.

IE if company X and company Y both quote 150CFM at the same head of pressure but company X's head has a 100cc port whereas company Y's has a 120cc port then comany X's head is unquestionably the one to go for.

Obviously the port volume an engine can 'live with' is dependant on a number of factors, the main ones being engine capacity and compression ratio. Revs is also a big factor but is probably more of result than a cause (hence later 4age heads with their much smaller ports produce similar power figures than the earlier ones).

I suppose that also goes for valve sizes.

One good thing the companies information does stress is it is the 'average' flow ("area under the graph") that is most important not just one single solitary peak figure. So, totally agree with you there.

That's why my head is so good despite a fairly low flow figure, it makes that figure from around 200 thou lift.

[greenamex2]

Oh yeah, further down the pages there are the actual flow numbers, may help rather than try and read the graphs.

[greenamex2]

And another thing I have just noticed.

Make sure that you know what Bore adapter diameter is being used.

CNC list theirs on the flow figures sheets. They use 100mm.

If another company used a smaller diameter then it may start to mask the valves and lower the flow figures. It doesn't mean their head is worse it just means that if you don't have a 100mm bore the lower figure is more likely to be much closer to reality

For example the Duratec has a 87.5mm bore so 100mm is well in excess of this. Looking at the pictures on the web site this engine has such huge valves as standard that they are going to get masked really easy at 87.5mm.

[GORDON STREETER]

Dave Moutain told me that the Duratec engine is the best thing that Fords have done for years. So is this the next engine that will be used for clubbie racing for the next 45 plus years like the good old Kent ?

[greenamex2]

The Kent engine had the benefit of being cheap to tune because it wasn't very good.

Looking at the Duratec, as soon as you go to unrestricted inlet, cams and exhaust it's going to produce a lot of horsepower, rev quite high, be highly stressed and need top quality (expensive) components.

Possibly not what the more budget orientated guys want.

Blinding for the rest of us though!

[James Murray]

How much more hp can be obtained by using the downdfraft head on the pre crossflows such as yours Zef? Was 120hp obtained in late 60's in 1000cc f3, or is that paddock gossip.

[knighty]

yes - the Duratec is a good motor - Dave Mountain would say that - he sells all the tuning parts for them........dont get me wrong - i was previously a design engineer at for Dave at Mountune and have done a lot of work on the Duratec - my honest opinion is the duratec head is certainly nothing special - actually quite a compromised head design - as its designed to work on a 1.8, 2.0 and 2.2 block, the ports are so big they actually have to run tumble flaps to speed up the port velocity.......flow is only part of the equation, you can fit big valves and port the head to hell and back and still not make much power........the trick with the duratec is to just give them a lick around the ports close to the valve seats.

port VELOCITY is the king when working on heads........the duratec ports also have the injectors mounted in the ports, which makes a real lash up of the port shaps - not good.........I have still yet to see anything that can rival a good old Vauxhall XE head,

with an intake runner inner diameter of 42mm , straight intake aimed at the back of the valves, 320mm total intake length, with a 125mm trumpet with a 7 degree inclusive angle, and a pair of 35mm intake valves - this is quite enough to produce over 270bhp.......this is what we did on the BTCC engines for Peugeot and Proton........big is not better.......but smaller is faster!

[knighty]

I also forgot to say - the trick when designing a port, or porting a head is to maintain attached flow.......in otherwords the flow must stick to the walls of the port........the injector bosses in the duratec ports will detach the flow and the flow will just tumble/bounce about in the port when the port harmonics take place - detached flow is the last thing you want.......Peugeot and Ranault dont do this, like the XE they are also good heads!.......if I was doing a balls out duratec race head, I would fill in the injector bosses with devcon / liquid ali putty stuff (not weld), and put the injector much further back, all in an attempt to keep attached flow........hope that helps

[greenamex2]

Originally Posted by knighty

the ports are so big they actually have to run tumble flaps to speed up the port velocity

the duratec ports also have the injectors mounted in the ports, which makes a real lash up of the port shaps - not good.........

Didn't ANYBODY learn from the Toyota 4a-ge? There was a good reason why subsequent variants of the engine removed these issues, and that was 15-20 years ago!

[phoenix]

Originally Posted by Dennis.Doyle

At tiny lifts, say 50 thou, the Duratec is flowing twice as much as the xflow.
By max lift, ~500 thou, it's only flowing 50% more.

I think it illustrates nicely the difference of opinion between Pheonix & Notso. At the tiny lifts, Phoenix's view that the effective area for the gas to get in is a cylinder-shaped "curtain" defined by the valve, the port it's just lifted off & the curved surface joining them up. The differences between those areas on the 2 & 4 valve engines is absolutely massive at the low lifts.

I'm not sure that I have a different opinion to Notso - as I haven't formed one yet!

What is clear to me though is that when a valve is lifted to 1/4 of its diameter (e.g. a 40 mm valve is lifted to 10 mm) the curtain area is equal to the area of the valve - about 12.56cm^2 in this case. As the area of the valve is always bigger than the area of the port (otherwise the valve would 'fall in' to the port!) it would seem that once the lift exceeds this, the open area of the valve will exceed the area of the port, so the port becomes the most significant restriction to flow rather than the valve.

Substituting two inlet valves of the same total area (28.28mm diameter) would give an improvement in flow of 141.42% of the single valve flow, upto about 7mm lift when the flow should 'max out'. Over the next 3mm of lift a single valve would close the gap right down, eventually achieving the same flow (in theory) as the two smaller valves.

I believe it is desirable to open the inlet valve as fast as possible; it must be easier to quickly accelerate to smaller valves to a peak lift of 7mm than a larger valve to 10mm of lift over the same number of crank angles?

But valves don't only flow at peak lift, and inlet valves spend much more time at partial lift than they do at full lift, so 'the area under the curve' we might be interested in is how much time is spent at 70% valve lift or less, for example. This will vary from cam to cam and engine to engine, but I reckon that in most cases the inlet valve is less than 70% open for 3 times as many crank degrees as the valve is open more than 70%. So if more than half the cylinder filling is done when the lift is less than 70% it would seem that power is to be gained if flow can be improved at lower lifts - which is what 4 valve heads seem to do very well!

Just a couple of things that I noticed on the CNC ported Duratec 2 litre head: The improvement seems to have been made by making the inlet ports smaller (2.24 square inches) rather than larger. With the standard size being given as 2.9 square inches this represents a REDUCTION to 77% of the standard port area. At the same time, the throat diameter is shown as increased by around 2% from 1.23 inches to 1.259 inches - giving an increase in throat area of 4.7%.

[greenamex2]

Err, how do you reduce the inlet port area, especially when the average port diameter has jumped from 1.23 to 1.259?

Have they welded in dividers?

This is something I've always though of doing but been advised against.

[phoenix]

Originally Posted by Denis Bassom

Err, how do you reduce the inlet port area, especially when the average port diameter has jumped from 1.23 to 1.259?

Have they welded in dividers?

This is something I've always though of doing but been advised against.

Port area has reduced from an average of 1.922" diameter to an average of 1.689" diameter. The average THROAT diameter has increased from 1.23" to 1.259". I guess they either weld up the ports then re-machine, or they use a 'filler' putty as knighty suggest previously, then re-shape.

[greenamex2]

Can you leave the filler putty in when using the engine or is it just for nice graphs/figures?

[phoenix]

Originally Posted by Denis Bassom

Can you leave the filler putty in when using the engine or is it just for nice graphs/figures?

I believe, if it's the right stuff, you leave it in. It bonds to the head and becomes a part of the head.

[GORDON STREETER]

You can use the two mix putty in the inlet ports as I have done with no problems . It also repairs the inlet port when you have ground into the water jacket .

[phoenix]

I have read that Loctite Metal Set A1 is the stuff to use on alloy heads - is that right anybody?