[fgc]

Hi

I just got a spare head for my nord 1750 and I am curious about the effect of porting. I have both of Jim K's books for the Alfa and I was super tempted to do everything accordingly.

However, this thread is not to call into questioin Jim's work as I think he did a wonderful job, just wondering if there are more than one way to Rome, so to speak, besides enlarging the ports.

Jim's work and the increase in power speaks for themselves. Just wondering about something that I read about (autodelta??) narrowing the port dimension to speed up the charge. Why so? Seems kind of counter intuitive and yet makes sense at the same time. It seems to me that they are trying to rely on the venturi effect to speed up the charge. (I am hardly an engineer so I cannot comment other than to read with a little common sense)

Anyone with experience narrowing ports or following this process? Sorry my memory about whether it was Autodelta that did this or the Competition manual etc I cannot remember.Just that I remembered reading somewhere about narrowing of the ports from the earlier to the later version. Comments will be most appreciated.

Again, I think there are probably more than 1 way to Rome and out of curiosity I would like to learn a little more about alternative ways of doing things apart from what is there since Autodelta apparently tried the same. Just exploring before I make the jump!

The advantage of taking Jim's route is that he has done everything so I only have to follow the text book. whereas anyother way I will have to figure some bits of it out. Anyway that is the fun bit about tinkering

Thanks
fgc

 

[George Willet]

The ports on the engine were TOO large, and port velocity suffered, so Autodelta made up some sleeves to bring the port size into line with what worked. Later heads had even smaller ports.

Shaping the port is more effective than enlarging the volumn.

 

[fgc]

Thanks George for your input.

I am wondering if Autodelta, whom I suppose must know their stuff , takes to stuffing the port to narrow it, why does it seem as if this practice is so uncommon in the power mod scene?

The common understanding about porting is quite different from what Autodelta discovered and also what I understand is being practiced by some manufacturer where their later offerings of updated engine in the same capacity showed smaller port. (that could be a red hearing but anyway)

Will appreciate if anyone has any dimension or diagrams on how Autodelta did their mod to narrow the port dimensions.

Thanks!
fgc

 

[slyalfa]

if you google you will see where this is done a lot on bikes. some of there ports were way way too big.
they fill it with epoxy and reshape it. A big port with a bad shape can be very bad. a small port with a very good shape will be better as long as the air speed is not too fast. I think the speed needed to stay under the speed of sound.

 

[George Willet]

FGC, I can fax the pages on the autodelta process, or mail copies to you out of the Compitition Advisory Book. Don't think I can put them on the BB. Maybe someone else can scan and paste them for us.

 

[180OUT]

The common understanding about porting is quite different from what Autodelta discovered and also what I understand is being practiced by some manufacturer where their later offerings of updated engine in the same capacity showed smaller port. (that could be a red hearing but anyway)

George is right, of course. I have the CAS . . . somewhere . . .and the diagram showing the small port mod on large port heads is certainly in line with the changes in design orthodoxy going on in the mid 60's. Generally speaking, smaller ports were found to produce more torque _and__power while large ports produced power, but only at high RPM's. Late model '67 GTV's for instance had substantially smaller ports than earlier 1600 GTV's.

JimK developed his port designs on a flow bench and then published the results. He's not just spouting bs. On the other hand, there are other approaches to porting Alfa heads (the stage 5 Mike Sperry head comes to mind) and several talented people here in the states (and elsewhere) are doing good work. Each one will have a slightly different approach and some, of course, work better than others. Jim is the only one I know of that has published the dimensions of his design, however.

Here's some homework for you. There are some interesting threads on this site regarding head porting. Do a search for them and I think you'll find them interesting. In addition to JimK's books, I suggest you read everything you can find on porting that's written by David Vizard. He's a very good technical writer and can explain the nuances of laminar flow and porting techniques better than just about anybody. In particular, you should try to get a copy of "The Theory And Practice Of Cylinder Head Modification".

I just noticed that Vizard is now the technical writer for Go Fast News. He has a very good article on that site (just read it tonight) in which, among other things, he points out why 4 valve heads don't produce as much torque as 2 valve heads. Lot's of people scratched their heads when Alfa came out with a twin spark head that 'only' had 2 valves per cylinder. Vizard's article explains why this is, in fact, a good idea. If you do a Google for David Vizard you'll had he has quite a lot of stuff on the 'net.

Oh, and check out the Alfaholics site. Max has a ported/modified head w/ 46x8 mm intakes, etc. that produces plus 40hp w/headers.

 

[Gordon Raymond]

History & new ideas

Some of the early work on Alfa ports was done by Horst Kwech and Ron Neal at Ausca. This information was passed on to Autodelta. Ausca's experimental work went further than most realize, with tapered ports and sleeves of different tapers and lengths to essentially move a secondary venturi closer to the port. This was before the days of flow benches and is a lot like Fords tunnel port heads of the era. In the day, Ausca's attempts were by-guess & by-gosh, trial and error. There never was a final solution as cam variables were thrown in the mix. Top end only power did not win races. Some flexibility is required. Autodelta did learn, large ports did not work as well as smaller, tapered ports, and Alfa head castings quickly changed over time. Today, with flow benches, and many other ways of measuring changes in valve pockets, port angles and shapes, combined with the effect of different cams, there are many new and better designs available. While my old copy of CAS does go into detail of how to modify large port GTA heads for as much flexibility as was expected of a race engine of the time, Jim k's more modern approach has the benefit of current technology in measuring gas flow in the Alfa heads. For a much more direct view, I am very interested in Richard Jemisons comments in regards to cam design in conjunction with Alfa valve pocket, valve size, and port runner design. Perhaps Alfar7 will give his opinion of porting possibilities with more current cam designs? fgc, don't hesitate to send him a PM about your post!
Gordon Raymond

 

[bearbvd]

The real problem with Alfa (nord) ports is the horrifically tight 'short side' radius. Jim K addresses this with his work.

The Ausca/Autodelta sleeves addressed it in a different way--if the port is smaller, PARTICULARLY if you hold the sleeve to the TOP of the port (which Ausca did NOT do), the short side radius gets longer.

What's going on is that the short radius on the short side causes a flow separation--and the turbulent burble downstream of that point makes the EFFECTIVE size of the port and the bowl much smaller than it appears to be.

A bigger port with a gentler short side radius will (generally) work best.

Ideal intake port air speed is around 300 fps, nowhere NEAR Mach 1.

Jim K touches on the REAL solution to Alfa ports--read the part about what happened to the exhaust port flow when some clay was placed in the bottome of the port, so as to give it a D shape---no real reason why this added material can't be aluminium added with a TIG torch, and, YES, it would also work on the intake side.

Greg

 

[Gordon Raymond]

Thanks Greg!

"The Ausca/Autodelta sleeves addressed it in a different way--if the port is smaller, PARTICULARLY if you hold the sleeve to the TOP of the port (which Ausca did NOT do), the short side radius gets longer.
What's going on is that the short radius on the short side causes a flow separation--and the turbulent burble downstream of that point makes the EFFECTIVE size of the port and the bowl much smaller than it appears to be." >GREGS QUOTE<

Greg,
Having seriously done your homework, you may also know that Ron N. (Ausca) addressed this exact issue by completely reworking an Alfa manifold and head {perhaps 1750 by then} using his "tubes" at a much higher angle through the manifold and head to straighten out the curve before the bowl. The welding, machine work and patching were amazing! By getting the runners way up there, the intake track was not only round, but straighter to the bowl. This worked well on the dyno, but was way to serious an alteration at the time, for the track. Little remained of the stock manifold. Ron's research did appear with the later GTAm's.
Your input is exactly what I have always believed.
Gordon Raymond

 

[genericwood]

no real reason why this added material can't be aluminium added with a TIG torch, and, YES, it would also work on the intake side.Greg

I'm convinced that the only reason more work hasn't been done on this is that SCCA production car rules allow removing material, but disallow adding material. So a lot of the more serious porting work was limited in its' ability to address the short-side radius Greg was referring to. As to reducing the overall turn the gases have to make, there are a lot of race engines where the top of the port has been moved up enough that the intake manifolds mounting holes have to be enlarged to slide it up higher on the head.

Ultimately, it seems way too much attention is normally given to the size of the ports, and too little to the shape.

Erik

 

[turbolarespider]

One of the things that you all have to bear in mind is that the head is a complete system- intake, port, valve, seats, cam profile, etc.

For the most part, the most important part is to match the porting job with the cams- as a massive generalization- high port velocities (smaller ports) match well with longer duration cams, and the opposite is generally true, too.

Big ports + big cams generally make an undriveable car....

Anyway, some of the original limitations are cam/ bucket material restrictions- with the materials on the cams and buckets you could get a certain maximum ramp rate- which would then pretty much dictate the lift and overlap all at once. With better cams, you can run more aggressive cam lift rates, and with totally different cams, get almost similar total open area/time.

The moral of the story- it's a system, and always should be considered as such- you pick the parts, and then choose the porting based on those parts.

Eric

 

[Alex]

Greg, Erik, Gordon, anyone!

Are you able to find and post diagrams, illustrations, photos etc of these ideas? In the past, I've only seen the briefest mention of the issue of tight port radii; and correct me if I'm wrong, but didn't someone once produce D-shaped exhaust manifolds to match a reworked port? I've seen a couple of bad photos of a Nord inlet port which had been reprofiled along one section using some kind of aluminum-rich putty (or JB Weld etc - what would be the best material to use??) but my memory of those is hazy. Instinct tells me that a D-shaped inlet port (as suggested by Greg) might require work to the inlet manifold too. Maybe Erik is right in that development work in this area (in the USA, at least) has been stymied by SCCA rules - but what about work done in Europe and elsewhere?

For anyone wanting to rework a head for the street, this looks like it's going to be a really informative thread.

Alex.

 

[genericwood]

I've seen a couple of bad photos of a Nord inlet port which had been reprofiled along one section using some kind of aluminum-rich putty (or JB Weld etc - what would be the best material to use??) but my memory of those is hazy. Alex.

Alex, as noted before, the right way to add material to the port is with TIG welding. As good as JB Weld and similar products are these days, I wouldn't use them for this application. Consider what would happen if a piece of JB Weld broke loose and lodged between the valve and seat.

I wouldn't undertake any of these efforts without a flow bench. A port that looks good won't necessarily flow. You really need to build it to a known spec (e.g. Jim K) or spend time on the bench.

Erik

 

[andyb6]

Not to add confusion to the topic,but.... Most engine builders will also tell you that the flow bench results don't always transfer to the dyno sheet. I had read somewhere a number of formula's to calculate an engines performance based on displacement and intake port size etc. The interesting thing was that through the magic or porting it is possible to make a naturally aspirated motor operate at OVER 100% efficiency. The idea being that the velocity of the incoming charge could create a positive pressure charge just before the valve snaps shut. So the magic is to find the harmonics of the intake charge and create the perfect intake speed at a given rpm range.
My race engines are currently operating at only 95% but over a wide range and I am running a relatively small port for a race engine. I know of some other huge port motors that have made more power but are actually less efficient and the efficiency is not flat.

I think the intake efficiency/speed/port size is probably best learned from our NASCAR buddies running restrictor plates. Unfortunately small block, push rod V-8 data does not directly transfer to our motors.

The more I think about it the more sense it makes and the less I understand it!

 

[Jim K.]

Nice thread you guys got going here! Just a few notes: Most race class rules preclude adding material to the head, but they don't mention the intake manifold! I say this, as if someone wants to create an obviously more effective 'downdraft' port, they can remove all the metal they want until they get too close to the follower well. Then, they can use a specially modified intake manifold with 'tongues' sticking out to fill-in the bottom of the head ports. This trick has been extensively used in F3 where open space allows downdrafting the heads in this way.
Regarding engine VE, its common to have sports/race engines with VE's around 103-107%. My engine simulator program puts my current project V6 engine at over 105% for a considerable part of the rev-range.
As for D-ports: Over the last 2 months, I dealt a bit with making 105 2liter D-shape intake ports through alloy welding and got about 128cfm with 46mm valves. The work was far from perfect, done in double time just to see if the car would perform better and it did. We didn't have the proper electrodes to properly get inside the ports so it looks a bit crappy. No power has been measured yet but we're working on that. One thing we didn't do is optimize valve shape for these ports. Its really strange to see a D ported/welded manifold (2x45DHLA). I should have taken a pic of the gasket we made! Comparable work was done in the exhaust ports with commensurate gains. Keep in mind that we had to make a special exhaust manifold to follow the redirected flow! Having gone through this, I'd say it's not worth the effort, especially since there are high-flow heads out there like the ones available from V.Sharp. Sit back, have a good time and let someone else who's real good at it, do the legwork!
Jim K.

 

[Alfar7]

Porting/valves

I'd say it's not worth the effort, especially since there are high-flow heads out there like the ones available from V.Sharp. Sit back, have a good time and let someone else who's real good at it, do the legwork!
Jim K.

I agree particularly when the cost/work results in nearly nothing as an improvement. Since there is no significant return and we only race for fun, reasonable modifications only seem to make sense. Overly large intake runners & ports generally fall in that area. Although the attached excerpt from a long email I sent to La_Strega_Nera was answereing his questions, I relayed some of my porting advice & am attaching that portion.

There are limits of flow in most small motors that limit how much valve & lift result in meaningful gains. Cams with over 10.5mm in a 1600/1750 is wasted wear & tear unless you are going for a 9000 rpm rebuild monthly motor. Same with exceeding 12.5mm on a 2 liter. There is just not enough flow through the port, bowl & seat to justify the additional drag and wear. I always raced on a budget & worked hard to make modifications cost effective. Older design cams and matched "sets" of same lift, and ramp rate cams for intake & exhaust leave horsepower and more importantly, TORQUE, going out the exhaust. Your cam profiles need to match the rest of the system. You can`t just plop a cam on a motor & expect great results. How commercial vendors can offer one or two cams & claim they are going to be really effective isn`t realistic. And they don`t give you the profile information to make a reasonable decision.
Off my soap box, here`s what I sent Ben re; porting & valves, etc.:

But a few Ideas you can use in your planning that might help before you buy things that might not help your performance.

1) remember that the 2.5 is basically the equivelent of a 1750 in valve sizes, yet has the same intake runner size as a 3.0 liter motor, (which has similar valve & seats to a 2 liter) so you have adequate runner size for significant improvement. Don`t plan on anything over 7200 rpms with the 12V 6 cyl. Little to gain...abd only use 164 type rockers. Heavier but will not bend & break so easily.

2) Intake runners.The runner size works very well as is, cleaned up,on both the 2.5 & 3 liter. I`m talking about the shorter Milano/GTV6 design. The 164 has larger longer runners and will work ok down low but they hurt upper rpm because of lack of flow speed & pulse timing. stick with the stock runner & initial intake port in the heads. Clean up the inside of the runners to get casting flaws out, and port match after head is done. We have built larger runners (Also Shankle) but nothing useable proved out on the dyno. Unfortunately vendors take advantage of the "bigger is better" mentality particularly on the BB.

3) I have always used stainless or steel valves in my heads. Get the sodium cooled tube valves out. They will break, They cannot take high rpms with a performance cam & extra seating pressures of racing valve springs.

4) Valve size. As to valve size remember this: they have to be the minimum size that will seat on the outer edge of the seats. Otherwise the larger diameter is an impediment to flow into the combustion chamber. The seat should be taper cut from the very base to the seat area determined by the outer diameter of the valves. The valve should have a seat area of .040 on the intake and .060 on the exhaust (cooling) The seat in a 2L 4cyl is as large as can realistically be fitted, the V6 12Vs can use a larger seat but not much will be gained if the valve, seats & bowl area match optimally. A larger valve on the existing seat cut correctly works very well. The use of larger valves requires the seating area at the absolute outer edge of the seat. (no reserve metal for strength, and if using a 45 degree cut, it is just a matter of time until the seat cracks.) When setting up a head this way I use a 30 degree seat cut. With a 30 degree cut there is less expanding force on the seat and there is less valve recessing over use. Since the seat should now be cut away tapering from the valve seat area to the top of the valve seatbowl floor. With the reduced internal area it effectively is a larger diameter seat for better flow. If you cut the typical 3 angle cut on the seats it reduces the inside diameter of the seat that can be cut away. At speed, gasses (fluid) do not attach to those small "radius" cuts. Get `er open!

5) Porting. Where the real gain in flow is made is in the area of the bowl & to the sides & top of the last half of the intake runner in the head. CAUTION! DO NOT cut away any material from the bottom third of the runner to the seat. That small hump must remain to attempt to direct flow to the bottom & lower sides of the valve. Luckily flow in that area is less important. The bowl should be shaped on top to initially match the flow line of the seat and cut about 2mm deeper back through & past the location of the guide, and tapered to the inlet. It is along with the area to the side of the ports from the guide to the seat base that can result in great gains as you should end up with a strait direct path to the seat from the inlet. Finish with a coarse sanding flap wheel just enough to smooth the peaks of the burr cuts. Do not polish. About 90 percent of this work could be done without removing the guides if building for a "not full race" head.
This will take a steady hand working with several different shaped burrs with long shanks to get in from the long curved intake on the V6 heads. Keep burrs & surface wet with mineral sprits (or diesel, but "no odor MS is much nicer!) or your burr will clog up right up. Use burrs for aluminum, not cast iron.
6) Fitting guides. You want the guide & seal to be at least .080 deeper or more if using lifts over 12mm) in the spring pocket.
Polish down an old guide so tht it will fit in the head without binding, and use to mark the point where inserted to the correct depth in the spring pocket, it meets the top of your finished port. (are they all the same?) Cut the guide off at a 30 degree angle from that point. cut back tip to leave a .030 flat surface around the valve hole. Debur.
On the exhaust side do the same but mark the point on yout test guide at the point on the head closest to the valve. Cut at the same 30 degrees, but leave a flat of .040 around the valve hole.
Install the guides. Seat the valves.
Tjhe first pic is a partially ported bowl with stock guide before the guide was removed to finish. The second pic is of a finish ported head before the 30 degree cut made to seats and surrounding aluminum to unshroud the seats and eliminate extra height of exhaust seat to improve flow into the valve.
Hope this helps.

Rj

 

[Gordon Raymond]

WoW!
Now with Jim K and Richard J looking in everyone can learn something! This is great. Thanks guys!
Here's some pictures to back up Richard's point # 3. A sodium GTA exhaust I keep on my desk, just so I remember. I am now experimenting with Titanium valves in an old Ausca 1600 single plug. Gordon Raymond

 

[AlfaTaso]

Thanks for my new Desktop backround picture Gordon!

 

[Gordon Raymond]

That was special for Richard. I know he likes Mick!
Gordon Raymond

 

[Gabor K.]

Put it in another way: Mostly people who want to port their heads, have just one head to work on and its their first try. So what is the chance for success in such a scenario?No experience, no flowbench, but a lot of interest in trying.

I think that those few experts who can port heads with consistent good results in the beginning ruined a lot of heads until lessons were learned and results were consistent. So chances of getting a good result for an optimistic novice in this business must be rather close to zero.

I have also heard of experts on one type of engine unseccesfully trying to transfer their knowledge to another brand and getting a result that did not work!

Personally I have done some modifications on the runners of the V6 and other engine but I would let the last centimeters in the head up to the valves be worked on by experts who could show to long experience!