Flow Benches and head porting

[knighty]

register on www.mototuneusa.com and all will be revealed - dont be put off with the tacky website - he has gone a bit OTT with the flashy stuff - but believe me - this boy is VERY clever......

look at the section titled "super sonic ports" its backed up by hundreds of case examples of very pleased people......from memory the stuff is JB weld, and you need the really hack up the ports and clean them well before you stick it on.......

what he does is fills up all the dead volume in the lower area of the intake ports with clay on a flow bench, to the point where it "just" nudges the port flow down a little, and then replicates this with the JB weld on the finished head........

believe me - smaller is faster, velocity is the key, not flow.......not to mention he is probably inducing a shed load of tumble, which 4-valve heads love!

welding a head to such an extent would distort it very badly, although people do doit and get away with it - but its risky - liquid ali putty is the stuff to use. just needs to be applied correctly.

[Dennis.Doyle]

Thanks for that guys - plenty to ponder.

P.S. I think the spare "RS2000" I picked up on ebay actually had more humble origins. Which do you think sounds better:

Van Diemen Scorpio
or
Van Diemen Galaxy

Either way, I reckon it will be good for taking the rise out of the Cosworth Duratecs & TTS V8s.

[GORDON STREETER]

James Murray, in reply to post 40, Cosworths were getting about 120bhp out of the old downdraft but that was on injection. I looked into this setup on carbs but I found my sidedraft head was about the same . I think Cosworth might have had a little more money /equipment/ and knowhow than me in my converted chicken shed !

[Robyn Slater]

I,m actually in the process of converting a head to downdraught at the moment.
A couple of things that i,m undecided about are the best tube diameter to use(1300cc engine) and also the method of fixing the port tubes into the head casting.I,ve been thinking of using locktite 648.
Any suggestions?

[GORDON STREETER]

The original one's were brazed in presumably because back then there was'nt the bonding technology of today. I have seen a few and actually owned a bonded one that seemed fine. The Dan Cox Anglia has one that has run for about 16 years with different owners over the years

[Notso Swift]

Great discussion and thanks for the links Dennis D.
Phoenix, that low speed flow is what will help the cyl fill, hence a better VE/more torque, which is what I was trying to say before, though in a typically inelegant way, by saying a 4v will get a much better result from the same (max) flow,
Dennis D Like Phoenix I don't think our opinions are actually different! Just the approach.

Raising the port floor is a pain in the back side but does have good results, I have a head guy who I saw do it on a Fiat twin cam for one of his customers, he compared Std, (cleaned), conventioanlly hand detailed port, D ported, where you fill the base and take the equivilent out of the sides of the port, to maintain volume, without getting any bigger, machined std angle port, machined raised port and raised port made with an inset ali tube. Generally speaking the Tube and the raised ports got the best results accross the board, with the conventionally detailed port third, but a way behind the other two, especially at low lift but it did catch up at higher lifts
The problem with both the raised and Tube ports is that you need to start lookng a making/machining your intakes to match, something that will cost all of the gains otherwise

Knighty thanks for the info, quality stuff, I had heard that the Duratec is a great motor for getting an "easy 230 ish hp", but after that the motor requires mega dollars, everything internal needs doing, while the XE has a good basis to improve in steps despite the initially smaller valves (33.5 vs 35mm). I have also heard that the M16 Pug is a bit of an undiscovered gem of a motor, really underrated, so your comments are a reinforcement of that. The thing is in Aus I don't get a whole lot of exposure to some of these things, they are all available, however, they are not much more than a blip on the local scene! For instance XE's only came in Calibra's, most Vectra's had 2.2 "LE" motors
Duratec is in the Focus, Mazda3+6 and the Tribute, but mostly in 2.3 forms, we no longer get the Mondeo and these are all new cars with few in the wreckers
Bugger all Mi16 were sold in Aus, most were 1.9 and very few 2.0, after all the Pug cost 50% more than a Mondeo or Vectra and about the same as a local V8.
Jap motors we get a lot more of, and I can recommend anyone looking at a 3SGE, for instance, as a fantastic base for a motor for a club level racer, less so the SR20, which takes 240-50 hp no problems, but after that gets tough. We once did a B6 that rev’d 11000 on its standard (mod) crank that put out 240 hp, not bad for a 1600, but that cost mega $

[knighty]

yes - the duratec is good for an easy 230 bhp, after that its time for serious surgery to the head........if I was doing a full race version looking for 270bhp from a 2.0 litre, I would look strongly at reducing the inlet port area to the equivalent surface area of a 42mm ID circle at the inlet manifold face.......either by the liquid-ali-putty stuff described above, or via some sort of manifold that protrudes / spigots into the port. adding material is illegal in many forms of motorsport, but not all........you would need the services of a 5 axis CNC vertical mill, like the ones used by the CNC porting companies, plus a bit of CAD modelling, but it could certainly be done.

regarding the ports, the other majic area to pay close attention to is the port choke ID's, which is simply the valve seat ID's........you will need to use some fancy valve seat material (berylium copper/nyla-bron) for a pair of 35mm intake valve OD's, the choke ID should be 32.5mm.........for the exhausts, valve OD of 30mm, choke ID should be 27mm.......hope that helps.......on the inlet, you will see thats really pushing it, with minimal seat-to-valve contact, but the general trick is to use the samllest inlet valve, with the biggest choke ID you dare, all in order to keep the valves de-shrouded from the bore, and keep the flow and velocity as good as possible.

[trikesrule]

I'm with you knighty mototune man has found something really really good. I recently finished a head using his ideas etc. I found a truck load of torque. That'll go very nicely on a new engine next season......trikes

[knighty]

Originally Posted by trikesrule

I'm with you knighty mototune man has found something really really good. I recently finished a head using his ideas etc. I found a truck load of torque. That'll go very nicely on a new engine next season......trikes

trikes - was it a 2-valve head?......or a 4-valve head?......push-rod or OHC?.....aluminium or cast iron?......the theory is very well proven on 4-valve heads, but by his admission, not yet on 2-valve heads - I see no reason why it shouldnt work on a 2-valve

If you did it on a 2-valve head I'm very very interested to hear more?........what sort of gain did you see over your base-line power and torque figures?........did you use a flow bench to chase out all the dead volume? - or just go ahead with his proven guide?.......please tell me more

[dtype38]

Going back to the two-valve / four-valve comparison again (if I may). I understand the effect of the lift height and that the effective restriction area passes from the curtain area to the port area as the valve(s) increase in lift. The figures Dennis.D showed, however, seem to assume identical cam profiles. Would I be right in saying that with smaller, lighter valves more aggressive profiles can be used (higher opening and closing accelerations) even at the same maximum lift? Would this have been done for any of the comparisons?

[knighty]

Dtype38 - yes you are quite right - two small valves offer much more aggressive opening profiles compared to opening one bigger heavier valve....in short - drastically less inertia to overcome.......this gives more area under the complete lift curve, more area = more power........for a pair of 35mm intake valves the flow pretty much maxes out at 12mm of lift, which is about 30% of the valve OD as a guide........the trick with a good pair of race cams is to open the valves as quick as possible, hold it open, then close them as quick as possible........this is where the combined inetria of the valve, cotter and tappet all come into play.........this is why finger followere are standard fitment to all F1 engines, the reduction in inertia of a finger follower compared to a tappet is about 30% reduction........not to mention less friction.........some of the race cams I have seen have been very "square".........generally speaking your kicking the proverbial out of the valve train. this is also why titanium valves need to be titanium-nitrided in order to give them more wear resistance.........and full race spec tappets and finger followers are now DLC coated along with the cams........they really do take a pounding on a proper race motor.

[zefarelly]

some of the original DD heads where cast as ( broadspeed is an example ) but as Gordon said most were brazed.

It straightlines the inlet tract which makes a difference, it also frees up the side of the head for the exhaust manifold and ports

Ultimately I guess you can over develope until the bank manager steps in if your that desperate , ,

I think broadspeed rad 1300 DD's with 145 BHP in the late 60's I know Dan Cox and others are running 160+ at 1500 nowadays

heres my head figures (current race head)

http://www.throbnozzle-racing.co.uk/...ow%20bench.htm

its not intended for general publication . . .but we're all freinds here !

[Dennis.Doyle]

Originally Posted by dtype38

Going back to the two-valve / four-valve comparison again (if I may). I understand the effect of the lift height and that the effective restriction area passes from the curtain area to the port area as the valve(s) increase in lift. The figures Dennis.D showed, however, seem to assume identical cam profiles. Would I be right in saying that with smaller, lighter valves more aggressive profiles can be used (higher opening and closing accelerations) even at the same maximum lift? Would this have been done for any of the comparisons?

Knighty's answered the valve train bit - I was just about to start speculating about cams a bit more. Those graphs are just head flow comparisons of course with valve lift plotted across the bottom. First refinement would be to transform the graph by factoring in the cam profile so that we could plot crank degrees across the bottom, then we could do 2 things. First, as Phoenix I think suggested we could look at the relative amount of time above/below certain lifts & get a handle on how much flow at low lifts matters. We would, of course, have a famly of curves for each head - one for each cam. Looking at the areas under the graphs we could say which profile got the best out of a given head.

Second refinement would be to start varying the pressure used in the tests. As a first step you could measure the pressure drop the piston creates in relation to crank degrees & those are the pressures you want to use for the flow readings at the lifts relating to those crank degrees.

Third refinement - do it at different revs as the fluid is going to behave differently, going supersonic & all that stuff.

Fourth - instrument a running engine so you can see what heat/combustion coming back up the inlet port during overlap etc is doing to it all.

By this time we might be competitive with the F3 teams - but then I think we might be spending an equivalent amount of time & money

Still, fun to think about, now must get back to that supersonic port stuff.

[Notso Swift]

I have to run a restrictor on my car so I have had more than my fill of the limits imposed by air going supersonic

[Dennis.Doyle]

Originally Posted by zefarelly

...

heres my head figures (current race head)

http://www.throbnozzle-racing.co.uk/...ow%20bench.htm

its not intended for general publication . . .but we're all freinds here !

Zef, I hadn't realized you were a devotee of enormous inlet valves. Did you keep exhaust one pretty much standard & go for offset guides to squeeze them in?

Does this also explain the huge flow you get at low lifts?

BTW, I had a closer look at the tables on the CNC pages. As someone (sorry to lazy to find acknowledge the kind soul) on the thread suggested if you click them you can read the figures. If you do you'll see some anomalies. As was pointed out their alloy head was tested on a slave bore of 100mm (a bit naughty since xflow valves are normally shrouded by the cylinder walls) whereas the std head they tested on a slave bore of 80mm (a shade undersize). Next the top table shows a single 41mm inlet & the lower table shows 2 inlet vales of 35mm dia - (a duratec head?) - their alloy is xflow isn't 4 valves in the pictures. Think this explains where the 200-230 claim comes from. Bit sloppy, huge cellar of salt required - handle with care.

[James Murray]

Gordon and Zef thank you for your counsel!

Gordon the 1000cc F3 car of the late 60's werent injected as the rules stipulated sincle choke carb.

Were you reffering to the Cosworth SCA downdrfaft which used Lucas mechanical injection. This was an f2 motor and used a Cosworth designed alloy head producing approx 140 BHP according to the Cosworth book.

[GORDON STREETER]

No I was thinking about the pushrod downdraft 1000 screamer. Cosworth were the first to "claim" 100 bhp per litre, and I think they finished up with 120bhp on Tecalemit injection . Mind you as I am 60 next week I am probably getting a bit senile ( to much rock n roll )

[zefarelly]

[QUOTE=Dennis.Doyle]Zef, I hadn't realized you were a devotee of enormous inlet valves. Did you keep exhaust one pretty much standard & go for offset guides to squeeze them in?

Does this also explain the huge flow you get at low lifts?

39.6 and 34.5 are homologated sizes as of 1963, no offset and plenty of meat between

I have done a head with 41.5 inlets . . . .untested as yet, but again, no offset but a bigger bore ( 85mm ) loads of compression and 33.5 exhausts

[dtype38]

Knightly and Dennis.D, thanks for that. I like to understand all of the parameters, even if I'm not up the higher end maths or testing techniques. Dennis, looking at your list of things to consider as variables... would it be true to say that gas extraction efficiency also has a big influence on cylinder filling and inlet flow. Which would imply that exhaust valve efficiency, size, porting, and exhaust back pressure would affect inlet tract flow.

Another one I just thought of is that on a running engine, the momentum across the head must have an effect on cylinder filling (hence the consideration of inlet velocity being more important than area) and that this is a function of the power of the engine (or vice-versa). That would mean that improving the power of an engine by any means (compression, chamber shape, timing etc, would generate more mass flow, which would actually affect the inlet tract sizing/optimum efficiency...

Is there a sort of rule of thumb about what factors are first order effect, second order, insignificant?

If I'm babbling, feel free to say so.

[Dennis.Doyle]

Don't credit us with too much - I can't do the maths anymore, I was never any good with experients & I was ony theorizing. Even so, I think the answer is "yes"

Basically, I think the interesting bit in a lot of this concerns overlap &/or when valves are open when the piston is going in the wrong direction. Despite the fact that the most bog-standard engines imagineable make use of that I find very difficult to get a feel for how big the effects must be. The effects include the exhaust flow sucking the inlet flow after it; scavenging; once a flow has some momentum then it'll carry on going until some other force stops it; turbulence getting good mixing & keeping detonation at bay; venturis; changing densities; stand-off, flames coming out the exhaust etc etc etc. Which are the big effects & which are the small ones is what baffles all of us.

Re. small ports & high velocities, I'm an agnostic. In it's favour, first I'd cite Cosworth. Remember the famous story of when they went to Indy in the '60s, saw some Offy heads & quipped that they should have arrows in them to tell the gas which way to go? Sounds to me like they were small(ish) port men. A second reason in favour would be sizing Weber chokes - we all know they can be too big for their own good. Conversely, I once saw a real cutaway Renault F1 engine & the ports didn't look that small to me (******* huge would be a better description). I also can't see that high velocity on it's own explains a lot. If you want to maximize momentum (M*V) then more flow (Cubic Feet per Min) does that if you keep time constant.

Couple or 4 thougts on that mototuneusa site. I couldn't find his explanation for his D-shaped ports at all - my guess is it's to do with scavenging between exhaust & inlet and/or maximizung the flow round the unshrouded sides of the valves. Second, bikes are pretty low-tech compared to F1 (much less time & money). Third, he's done quite a lot on bhp-restricted stuff - hence the benefit is all in terms of wider power bands (useful in rally/hillclimb applications?). Fourth, bikes & riders have problems handling ginormous horepower at the best of times - he himself says smooth delivery matters (& it does to me too).

It's just a great big fog to me. Anyway, must stop warbling & see what trouble I've got into elsewhere.

[GORDON STREETER]

Perhaps Fords 2 stroke engine in my thread is the answer, as I remember reading about the old Honda 500 2 stroke GP bike engine "claiming" 380 bhp at a certain revs but had to be set to 200 bhp to make it rideable. Perhaps I'll see if it fits the Anglebox If I got in front you would'nt be able to see me!

[zefarelly]

slightly off topic, my first engine build was a 50cc 2 stroke Vespa when I was 15 . . .played with Lambrettas for 10 years bofore moving onto cars ( 90+mph on a LAmbretta makes racing cortinas seem fairly tame !)

one of my earliest races at Lyden was with a 2 stroke saab . . .I'd love to see how that engine works inside . . .if the owner had used better 2 stroke oil I might have seen the track better too !

[dtype38]

Ok, thanks Dennis. Understand where you're at and believe me, its at a higher place than me :-))

Zef.. did a bit of 2-stroking myself a while back. Had a Suzuki tripple with mechanical lift carbs and no filters. If you stuck your head against the oil tank, looked down an outer choke and opened the throttle, you could actually watch the piston going up and down!!! Facinating, but not something you'd want to watch for long after you start thinking about how the flow knows to go in the right direction :-)

[Dennis.Doyle]

Originally Posted by GORDON STREETER

No I was thinking about the pushrod downdraft 1000 screamer. Cosworth were the first to "claim" 100 bhp per litre, and I think they finished up with 120bhp on Tecalemit injection . Mind you as I am 60 next week I am probably getting a bit senile ( to much rock n roll )

Re. the downdraft heads. I'm sure there is a good old empirical formula for calculating how much energy you take from a flow in order to make it do a 90 left, but I'm rogered if I can find it. Anyway, the problems of subsequently keeping it gas, water & oil tight would put me right off the idea.

Senile? Can't remember the meaning of the word. Yesterday I had to say "country to the left of Spain". TRUE STORY.

Too much rock n roll == meaningless concept, but keeping up with new developments is definitely a challenge.

[Notso Swift]

Originally Posted by Dennis.Doyle

Senile? Can't remember the meaning of the word. Yesterday I had to say "country to the left of Spain". TRUE STORY.

Wouldn't that depend on which way you were facing